Unnamed: 0.1
int64
0
41k
Unnamed: 0
int64
0
41k
author
stringlengths
9
1.39k
id
stringlengths
11
18
summary
stringlengths
25
3.66k
title
stringlengths
4
258
year
int64
1.99k
2.02k
arxiv_url
stringlengths
32
39
info
stringlengths
523
3.18k
embeddings
stringlengths
16.9k
17.1k
40,400
40,400
['Daniel Golovin', 'Andreas Krause', 'Matthew Streeter']
0908.0772v1
Which ads should we display in sponsored search in order to maximize our revenue? How should we dynamically rank information sources to maximize value of information? These applications exhibit strong diminishing returns: Selection of redundant ads and information sources decreases their marginal utility. We show that ...
Online Learning of Assignments that Maximize Submodular Functions
2,009
http://arxiv.org/pdf/0908.0772v1
Title Online Learning Assignments Maximize Submodular Functions Summary ad display sponsored search order maximize revenue dynamically rank information source maximize value information application exhibit strong diminishing return Selection redundant ad information source decrease marginal utility show problem formali...
[0.0702047273516655, 0.036722730845212936, -0.02226448431611061, -0.04861723631620407, -0.006673960480839014, -0.02619578316807747, 0.035697754472494125, 0.005653332453221083, 0.006916348822414875, -0.07342896610498428, -0.028042521327733994, 0.0046054101549088955, 0.006591056007891893, 0.07499989122152328, -0.01421258...
40,401
40,401
['C. Balasubramanian', 'K. Duraiswamy']
0908.0984v1
In real life, media information has time attributes either implicitly or explicitly known as temporal data. This paper investigates the usefulness of applying Bayesian classification to an interval encoded temporal database with prioritized items. The proposed method performs temporal mining by encoding the database wi...
An Application of Bayesian classification to Interval Encoded Temporal mining with prioritized items
2,009
http://arxiv.org/pdf/0908.0984v1
Title Application Bayesian classification Interval Encoded Temporal mining prioritized item Summary real life medium information time attribute either implicitly explicitly known temporal data paper investigates usefulness applying Bayesian classification interval encoded temporal database prioritized item proposed met...
[-0.028508933261036873, 0.03028908185660839, -0.035837572067976, -0.027262991294264793, -0.030272556468844414, 0.0138474116101861, -0.006793288514018059, 0.0297352597117424, -0.0379491001367569, -0.03716377913951874, 0.07607296854257584, 0.008877974934875965, 0.0126156285405159, 0.056196145713329315, -0.065600447356700...
40,402
40,402
['Robert Kleinberg', 'Aleksandrs Slivkins']
0911.1174v1
The Lipschitz multi-armed bandit (MAB) problem generalizes the classical multi-armed bandit problem by assuming one is given side information consisting of a priori upper bounds on the difference in expected payoff between certain pairs of strategies. Classical results of (Lai and Robbins 1985) and (Auer et al. 2002) i...
Sharp Dichotomies for Regret Minimization in Metric Spaces
2,009
http://arxiv.org/pdf/0911.1174v1
Title Sharp Dichotomies Regret Minimization Metric Spaces Summary Lipschitz multiarmed bandit MAB problem generalizes classical multiarmed bandit problem assuming one given side information consisting priori upper bound difference expected payoff certain pair strategy Classical result Lai Robbins 1985 Auer et al 2002 i...
[-0.017507262527942657, 0.06138422712683678, -0.015130091458559036, -0.03270373120903969, -0.01688886433839798, -0.0056529720313847065, -0.01566641964018345, 0.03446073830127716, 0.003605370642617345, 0.030075138434767723, -0.022989902645349503, 0.005174126476049423, -0.05315357446670532, -0.009612328372895718, 0.04441...
40,403
40,403
['Shipra Agrawal', 'Zizhuo Wang', 'Yinyu Ye']
0911.2974v3
A natural optimization model that formulates many online resource allocation and revenue management problems is the online linear program (LP) in which the constraint matrix is revealed column by column along with the corresponding objective coefficient. In such a model, a decision variable has to be set each time a co...
A Dynamic Near-Optimal Algorithm for Online Linear Programming
2,009
http://arxiv.org/pdf/0911.2974v3
Title Dynamic NearOptimal Algorithm Online Linear Programming Summary natural optimization model formulates many online resource allocation revenue management problem online linear program LP constraint matrix revealed column column along corresponding objective coefficient model decision variable set time column revea...
[0.026490269228816032, 0.022129103541374207, -0.028243090957403183, -0.04874974489212036, -0.011181934736669064, -0.021900618448853493, 5.994938692310825e-05, 0.032957468181848526, -0.052246857434511185, -0.018154853954911232, -0.009805353358387947, 0.05471175163984299, -0.005365205463021994, 0.10882438719272614, -0.02...
40,404
40,404
['Iza Marfisi-Schottman', 'Aymen Sghaier', 'Sébastien George', 'Franck Tarpin-Bernard', 'Patrick Prévôt']
0911.4262v1
Serious Games (SGs) have experienced a tremendous outburst these last years. Video game companies have been producing fun, user-friendly SGs, but their educational value has yet to be proven. Meanwhile, cognition research scientist have been developing SGs in such a way as to guarantee an educational gain, but the fun ...
Towards Industrialized Conception and Production of Serious Games
2,009
http://arxiv.org/pdf/0911.4262v1
Title Towards Industrialized Conception Production Serious Games Summary Serious Games SGs experienced tremendous outburst last year Video game company producing fun userfriendly SGs educational value yet proven Meanwhile cognition research scientist developing SGs way guarantee educational gain fun attractive characte...
[0.030101409181952477, 0.0026686203200370073, -0.04647606611251831, -0.005047663114964962, 0.014154067263007164, -0.031771089881658554, 0.056515686213970184, -0.011564163491129875, -0.030035672709345818, 0.0013144179247319698, 0.01356200035661459, 0.05179426819086075, 0.010118752717971802, 0.12463058531284332, 0.028442...
40,405
40,405
['Hartmut Neven', 'Vasil S. Denchev', 'Geordie Rose', 'William G. Macready']
0912.0779v1
In a previous publication we proposed discrete global optimization as a method to train a strong binary classifier constructed as a thresholded sum over weak classifiers. Our motivation was to cast the training of a classifier into a format amenable to solution by the quantum adiabatic algorithm. Applying adiabatic qua...
Training a Large Scale Classifier with the Quantum Adiabatic Algorithm
2,009
http://arxiv.org/pdf/0912.0779v1
Title Training Large Scale Classifier Quantum Adiabatic Algorithm Summary previous publication proposed discrete global optimization method train strong binary classifier constructed thresholded sum weak classifier motivation cast training classifier format amenable solution quantum adiabatic algorithm Applying adiabat...
[-0.024085475131869316, 0.047445010393857956, -0.032188743352890015, 0.02396136149764061, -0.008128035813570023, 0.0025820264127105474, -0.060947660356760025, 0.04676235094666481, 0.010696201585233212, -0.024722987785935402, -0.02952873706817627, -0.033422183245420456, -0.04385358840227127, -0.0035078609362244606, 0.02...
40,406
40,406
['Shailendra Singh', 'Sanjay Silakari']
0912.1014v1
Feature selection is an indispensable preprocessing step when mining huge datasets that can significantly improve the overall system performance. Therefore in this paper we focus on a hybrid approach of feature selection. This method falls into two phases. The filter phase select the features with highest information g...
An ensemble approach for feature selection of Cyber Attack Dataset
2,009
http://arxiv.org/pdf/0912.1014v1
Title ensemble approach feature selection Cyber Attack Dataset Summary Feature selection indispensable preprocessing step mining huge datasets significantly improve overall system performance Therefore paper focus hybrid approach feature selection method fall two phase filter phase select feature highest information ga...
[-0.003591942135244608, 0.02275639772415161, -0.04871070012450218, 0.017847741022706032, -0.032034892588853836, -0.02174791879951954, 0.08186332136392593, -0.009611387737095356, 0.020986344665288925, -0.04730423912405968, 0.07669340074062347, 0.0073052383959293365, -0.0034109707921743393, 0.0665183961391449, -0.0090074...
40,407
40,407
['Daniel M. Kane']
0912.2709v1
We provide asymptotically sharp bounds for the Gaussian surface area and the Gaussian noise sensitivity of polynomial threshold functions. In particular we show that if $f$ is a degree-$d$ polynomial threshold function, then its Gaussian sensitivity at noise rate $\epsilon$ is less than some quantity asymptotic to $\fr...
The Gaussian Surface Area and Noise Sensitivity of Degree-$d$ Polynomials
2,009
http://arxiv.org/pdf/0912.2709v1
Title Gaussian Surface Area Noise Sensitivity Degreed Polynomials Summary provide asymptotically sharp bound Gaussian surface area Gaussian noise sensitivity polynomial threshold function particular show f degreed polynomial threshold function Gaussian sensitivity noise rate epsilon le quantity asymptotic fracdsqrt2eps...
[-0.06717288494110107, 0.0016383797628805041, -0.0322284996509552, 0.08547835797071457, 0.035850897431373596, -0.02911842241883278, 0.04814301058650017, -0.0281640887260437, 0.017811670899391174, 0.00863207969814539, 0.02364012785255909, 0.02557387202978134, 0.016720090061426163, 0.058879293501377106, -0.02958033978939...
40,408
40,408
['O. J. Oyelade', 'O. O. Oladipupo', 'I. C. Obagbuwa']
1002.2425v1
The ability to monitor the progress of students academic performance is a critical issue to the academic community of higher learning. A system for analyzing students results based on cluster analysis and uses standard statistical algorithms to arrange their scores data according to the level of their performance is de...
Application of k Means Clustering algorithm for prediction of Students Academic Performance
2,010
http://arxiv.org/pdf/1002.2425v1
Title Application k Means Clustering algorithm prediction Students Academic Performance Summary ability monitor progress student academic performance critical issue academic community higher learning system analyzing student result based cluster analysis us standard statistical algorithm arrange score data according le...
[0.012255260720849037, -0.03238091245293617, -0.05472239851951599, 0.01160482782870531, 0.017057953402400017, 0.02625643089413643, 0.02851349115371704, 0.006601156201213598, 0.03528141602873802, -0.016666606068611145, 0.03313330188393593, 0.054519180208444595, 0.016980869695544243, 0.05541977286338806, -0.0089536160230...
40,409
40,409
['Vitaly Feldman']
1002.3183v3
Statistical query (SQ) learning model of Kearns (1993) is a natural restriction of the PAC learning model in which a learning algorithm is allowed to obtain estimates of statistical properties of the examples but cannot see the examples themselves. We describe a new and simple characterization of the query complexity o...
A Complete Characterization of Statistical Query Learning with Applications to Evolvability
2,010
http://arxiv.org/pdf/1002.3183v3
Title Complete Characterization Statistical Query Learning Applications Evolvability Summary Statistical query SQ learning model Kearns 1993 natural restriction PAC learning model learning algorithm allowed obtain estimate statistical property example cannot see example describe new simple characterization query comple...
[-0.021647272631525993, 0.03767754137516022, -0.018612079322338104, -0.011136188171803951, -0.04832536354660988, -0.013785481452941895, 0.00023721138131804764, 0.018488900735974312, -0.03559328615665436, -0.004494735971093178, 0.020736107602715492, 0.0024252310395240784, -0.016100913286209106, 0.06379442662000656, 0.00...
40,410
40,410
['G. Nathiya', 'S. C. Punitha', 'M. Punithavalli']
1004.1743v1
Clustering is an unsupervised learning method that constitutes a cornerstone of an intelligent data analysis process. It is used for the exploration of inter-relationships among a collection of patterns, by organizing them into homogeneous clusters. Clustering has been dynamically applied to a variety of tasks in the f...
An Analytical Study on Behavior of Clusters Using K Means, EM and K* Means Algorithm
2,010
http://arxiv.org/pdf/1004.1743v1
Title Analytical Study Behavior Clusters Using K Means EM K Means Algorithm Summary Clustering unsupervised learning method constitutes cornerstone intelligent data analysis process used exploration interrelationship among collection pattern organizing homogeneous cluster Clustering dynamically applied variety task fie...
[0.02352573536336422, -0.056265104562044144, -0.05099105089902878, 0.02970346249639988, -0.007772560231387615, 0.018338309600949287, 0.03691679239273071, 0.04018562287092209, 0.02192329429090023, -0.005122468341141939, -0.002322238404303789, 0.05109986290335655, 0.02068428508937359, 0.011929074302315712, 0.004583447705...
40,411
40,411
['Ankur Moitra', 'Gregory Valiant']
1004.4223v1
Given data drawn from a mixture of multivariate Gaussians, a basic problem is to accurately estimate the mixture parameters. We give an algorithm for this problem that has a running time, and data requirement polynomial in the dimension and the inverse of the desired accuracy, with provably minimal assumptions on the G...
Settling the Polynomial Learnability of Mixtures of Gaussians
2,010
http://arxiv.org/pdf/1004.4223v1
Title Settling Polynomial Learnability Mixtures Gaussians Summary Given data drawn mixture multivariate Gaussians basic problem accurately estimate mixture parameter give algorithm problem running time data requirement polynomial dimension inverse desired accuracy provably minimal assumption Gaussians simple consequenc...
[-0.045064058154821396, 0.052726563066244125, 0.0033657762687653303, 0.031998466700315475, 0.021868102252483368, 0.007634799927473068, 0.025891361758112907, 0.017408568412065506, -0.0223164614289999, 0.03955914080142975, 0.015882831066846848, 0.02443012036383152, 0.02571365237236023, 0.030921481549739838, 0.01905695162...
40,412
40,412
['Mikhail Belkin', 'Kaushik Sinha']
1004.4864v1
The question of polynomial learnability of probability distributions, particularly Gaussian mixture distributions, has recently received significant attention in theoretical computer science and machine learning. However, despite major progress, the general question of polynomial learnability of Gaussian mixture distri...
Polynomial Learning of Distribution Families
2,010
http://arxiv.org/pdf/1004.4864v1
Title Polynomial Learning Distribution Families Summary question polynomial learnability probability distribution particularly Gaussian mixture distribution recently received significant attention theoretical computer science machine learning However despite major progress general question polynomial learnability Gauss...
[-0.048455610871315, 0.013886814936995506, -0.012303175404667854, 0.022119494155049324, 0.0026823871303349733, -0.013536171987652779, -0.01847713813185692, -0.018582701683044434, -0.011211234144866467, 0.017825499176979065, 0.0573764443397522, 0.03624753654003143, 0.011723137460649014, 0.05797772482037544, 0.0055258776...
40,413
40,413
['Adam Gauci', 'Kristian Zarb Adami', 'John Abela']
1005.0390v2
In this work, decision tree learning algorithms and fuzzy inferencing systems are applied for galaxy morphology classification. In particular, the CART, the C4.5, the Random Forest and fuzzy logic algorithms are studied and reliable classifiers are developed to distinguish between spiral galaxies, elliptical galaxies o...
Machine Learning for Galaxy Morphology Classification
2,010
http://arxiv.org/pdf/1005.0390v2
Title Machine Learning Galaxy Morphology Classification Summary work decision tree learning algorithm fuzzy inferencing system applied galaxy morphology classification particular CART C45 Random Forest fuzzy logic algorithm studied reliable classifier developed distinguish spiral galaxy elliptical galaxy starunknown ga...
[0.01436183787882328, -0.018092073500156403, -0.007626781240105629, 0.0038823490031063557, 0.017313862219452858, 0.033840641379356384, 0.02383428066968918, 0.04926507547497749, -0.03772355243563652, 0.0015097421128302813, 0.008867201395332813, -0.026521293446421623, 0.03182337433099747, -0.007836838252842426, -0.013818...
40,414
40,414
['Sumit Ganguly']
1005.1120v2
For each $p \in (0,2]$, we present a randomized algorithm that returns an $\epsilon$-approximation of the $p$th frequency moment of a data stream $F_p = \sum_{i = 1}^n \abs{f_i}^p$. The algorithm requires space $O(\epsilon^{-2} \log (mM)(\log n))$ and processes each stream update using time $O((\log n) (\log \epsilon^{...
Estimating small moments of data stream in nearly optimal space-time
2,010
http://arxiv.org/pdf/1005.1120v2
Title Estimating small moment data stream nearly optimal spacetime Summary p 02 present randomized algorithm return epsilonapproximation pth frequency moment data stream Fp sumi 1n absfip algorithm requires space Oepsilon2 log mMlog n process stream update using time Olog n log epsilon1 nearly optimal term space lower ...
[-0.04558267444372177, 0.010708533227443695, 0.018347691744565964, -0.0022658915258944035, -0.018991196528077126, -0.03944000229239464, 0.006530310492962599, 0.024844704195857048, -0.07021108269691467, 0.03398754820227623, 0.07187191396951675, -0.03920904919505119, -0.0007672809879295528, 0.05744997039437294, -0.019592...
40,415
40,415
['Ankan Saha', 'Ambuj Tewari']
1005.2146v1
Cyclic coordinate descent is a classic optimization method that has witnessed a resurgence of interest in machine learning. Reasons for this include its simplicity, speed and stability, as well as its competitive performance on $\ell_1$ regularized smooth optimization problems. Surprisingly, very little is known about ...
On the Finite Time Convergence of Cyclic Coordinate Descent Methods
2,010
http://arxiv.org/pdf/1005.2146v1
Title Finite Time Convergence Cyclic Coordinate Descent Methods Summary Cyclic coordinate descent classic optimization method witnessed resurgence interest machine learning Reasons include simplicity speed stability well competitive performance ell1 regularized smooth optimization problem Surprisingly little known fini...
[-0.01904691383242607, 0.026750218123197556, 0.009517468512058258, -0.03203791007399559, -0.010934032499790192, 0.012588400393724442, -0.015067817643284798, -0.010718051344156265, 0.03577103838324547, 0.022457823157310486, 0.019103484228253365, -0.012376347556710243, 0.020017148926854134, -0.06230856850743294, -0.00098...
40,416
40,416
['Volker Nannen']
1005.2364v2
The concept of overfitting in model selection is explained and demonstrated with an example. After providing some background information on information theory and Kolmogorov complexity, we provide a short explanation of Minimum Description Length and error minimization. We conclude with a discussion of the typical feat...
A Short Introduction to Model Selection, Kolmogorov Complexity and Minimum Description Length (MDL)
2,010
http://arxiv.org/pdf/1005.2364v2
Title Short Introduction Model Selection Kolmogorov Complexity Minimum Description Length MDL Summary concept overfitting model selection explained demonstrated example providing background information information theory Kolmogorov complexity provide short explanation Minimum Description Length error minimization concl...
[-0.04456401243805885, -0.016315944492816925, -0.01774989254772663, -0.024861404672265053, 0.005788037553429604, 0.012675956822931767, 0.0740591362118721, 0.0015853088116273284, -0.0146133191883564, 0.02477426268160343, 0.057281434535980225, 0.04798082634806633, 0.04674388840794563, 0.05345752090215683, -0.023792073130...
40,417
40,417
['Andri Mirzal', 'Masashi Furukawa']
1005.2603v5
This paper presents a concise tutorial on spectral clustering for broad spectrum graphs which include unipartite (undirected) graph, bipartite graph, and directed graph. We show how to transform bipartite graph and directed graph into corresponding unipartite graph, therefore allowing a unified treatment to all cases. ...
Eigenvectors for clustering: Unipartite, bipartite, and directed graph cases
2,010
http://arxiv.org/pdf/1005.2603v5
Title Eigenvectors clustering Unipartite bipartite directed graph case Summary paper present concise tutorial spectral clustering broad spectrum graph include unipartite undirected graph bipartite graph directed graph show transform bipartite graph directed graph corresponding unipartite graph therefore allowing unifie...
[-0.02489628829061985, -0.08898556232452393, -0.021413160488009453, 0.03332250565290451, -0.035610780119895935, 0.011725828982889652, 0.015067119151353836, -0.015249006450176239, 0.039397530257701874, -0.011923347599804401, -0.046796634793281555, -0.0014346357202157378, 0.005358168855309486, -0.013457916676998138, -0.0...
40,418
40,418
['James P. Crutchfield', 'Sean Whalen']
1005.2714v2
We introduce a theory of sequential causal inference in which learners in a chain estimate a structural model from their upstream teacher and then pass samples from the model to their downstream student. It extends the population dynamics of genetic drift, recasting Kimura's selectively neutral theory as a special case...
Structural Drift: The Population Dynamics of Sequential Learning
2,010
http://arxiv.org/pdf/1005.2714v2
Title Structural Drift Population Dynamics Sequential Learning Summary introduce theory sequential causal inference learner chain estimate structural model upstream teacher pas sample model downstream student extends population dynamic genetic drift recasting Kimuras selectively neutral theory special case generalized ...
[-0.038304127752780914, 0.056641701608896255, -0.04565886780619621, -0.061517275869846344, 0.0006457430426962674, -0.03545159101486206, 0.01628267392516136, -0.01705172471702099, -0.010041468776762486, -0.0013420230243355036, 0.05282964929938316, 0.015454665757715702, 0.002656571101397276, 0.030599189922213554, 0.02678...
40,419
40,419
['Pierre-François Marteau', 'Sylvie Gibet']
1005.5141v12
This paper proposes some extensions to the work on kernels dedicated to string or time series global alignment based on the aggregation of scores obtained by local alignments. The extensions we propose allow to construct, from classical recursive definition of elastic distances, recursive edit distance (or time-warp) k...
On Recursive Edit Distance Kernels with Application to Time Series Classification
2,010
http://arxiv.org/pdf/1005.5141v12
Title Recursive Edit Distance Kernels Application Time Series Classification Summary paper proposes extension work kernel dedicated string time series global alignment based aggregation score obtained local alignment extension propose allow construct classical recursive definition elastic distance recursive edit distan...
[-0.034243084490299225, 0.026403073221445084, -0.016830019652843475, 0.019059762358665466, -0.046611763536930084, 0.003913916181772947, 0.015439056791365147, 0.06142328307032585, -0.023709937930107117, -0.02225969359278679, 0.0787755697965622, -0.009643875062465668, 0.03976677730679512, 0.039862338453531265, 0.00789438...
40,420
40,420
['P. Bouboulis']
1005.5170v1
The present report, has been inspired by the need of the author and its colleagues to understand the underlying theory of Wirtinger's Calculus and to further extend it to include the kernel case. The aim of the present manuscript is twofold: a) it endeavors to provide a more rigorous presentation of the related materia...
Wirtinger's Calculus in general Hilbert Spaces
2,010
http://arxiv.org/pdf/1005.5170v1
Title Wirtingers Calculus general Hilbert Spaces Summary present report inspired need author colleague understand underlying theory Wirtingers Calculus extend include kernel case aim present manuscript twofold endeavor provide rigorous presentation related material focusing aspect author find insightful b extends notio...
[-0.021877683699131012, 0.0376695841550827, -0.008522601798176765, 0.009805820882320404, -0.03079058974981308, 0.012405401095747948, -0.020567256957292557, 0.029915183782577515, 0.005981572903692722, 0.03404664993286133, -0.03742436319589615, -0.034957434982061386, 0.006946624256670475, -0.04005247354507446, -0.0011988...
40,421
40,421
['Aleksandrs Slivkins', 'Filip Radlinski', 'Sreenivas Gollapudi']
1005.5197v2
Most learning to rank research has assumed that the utility of different documents is independent, which results in learned ranking functions that return redundant results. The few approaches that avoid this have rather unsatisfyingly lacked theoretical foundations, or do not scale. We present a learning-to-rank formul...
Ranked bandits in metric spaces: learning optimally diverse rankings over large document collections
2,010
http://arxiv.org/pdf/1005.5197v2
Title Ranked bandit metric space learning optimally diverse ranking large document collection Summary learning rank research assumed utility different document independent result learned ranking function return redundant result approach avoid rather unsatisfyingly lacked theoretical foundation scale present learningtor...
[0.04623692110180855, 0.05102129280567169, 0.0015441037248820066, 0.004123441409319639, 0.02862010896205902, 0.006811399478465319, 0.01695304363965988, 0.045195113867521286, 0.04182892292737961, -0.027121152728796005, -0.0500151701271534, -0.00902519654482603, -0.05029799044132233, 0.009584684856235981, -0.000291672768...
40,422
40,422
['Yucheng Low', 'Joseph Gonzalez', 'Aapo Kyrola', 'Danny Bickson', 'Carlos Guestrin', 'Joseph M. Hellerstein']
1006.4990v1
Designing and implementing efficient, provably correct parallel machine learning (ML) algorithms is challenging. Existing high-level parallel abstractions like MapReduce are insufficiently expressive while low-level tools like MPI and Pthreads leave ML experts repeatedly solving the same design challenges. By targeting...
GraphLab: A New Framework for Parallel Machine Learning
2,010
http://arxiv.org/pdf/1006.4990v1
Title GraphLab New Framework Parallel Machine Learning Summary Designing implementing efficient provably correct parallel machine learning ML algorithm challenging Existing highlevel parallel abstraction like MapReduce insufficiently expressive lowlevel tool like MPI Pthreads leave ML expert repeatedly solving design c...
[-0.009661894291639328, 0.034596532583236694, -0.015733323991298676, -0.015012306161224842, -0.035003118216991425, 0.009322117082774639, 0.019773181527853012, -0.006333834491670132, 0.06950442492961884, 0.02598232589662075, 0.05510793253779411, 0.056912861764431, 0.0053227609023451805, 0.04073081910610199, -0.012754894...
40,423
40,423
['Madjid Khalilian', 'Norwati Mustapha']
1006.5261v1
Very large databases are required to store massive amounts of data that are continuously inserted and queried. Analyzing huge data sets and extracting valuable pattern in many applications are interesting for researchers. We can identify two main groups of techniques for huge data bases mining. One group refers to stre...
Data Stream Clustering: Challenges and Issues
2,010
http://arxiv.org/pdf/1006.5261v1
Title Data Stream Clustering Challenges Issues Summary large database required store massive amount data continuously inserted queried Analyzing huge data set extracting valuable pattern many application interesting researcher identify two main group technique huge data base mining One group refers streaming data appli...
[-0.026489216834306717, 0.0018132856348529458, -0.0533626489341259, -0.007504562847316265, -0.019378935918211937, 0.029120326042175293, -0.02886327914893627, -0.007601575925946236, 0.027643028646707535, 0.008396636694669724, 0.03078877553343773, 0.02353585697710514, -0.007751514203846455, 0.06709583103656769, -0.007721...
40,424
40,424
['Dhoha Almazro', 'Ghadeer Shahatah', 'Lamia Albdulkarim', 'Mona Kherees', 'Romy Martinez', 'William Nzoukou']
1006.5278v4
Recommender systems apply data mining techniques and prediction algorithms to predict users' interest on information, products and services among the tremendous amount of available items. The vast growth of information on the Internet as well as number of visitors to websites add some key challenges to recommender syst...
A Survey Paper on Recommender Systems
2,010
http://arxiv.org/pdf/1006.5278v4
Title Survey Paper Recommender Systems Summary Recommender system apply data mining technique prediction algorithm predict user interest information product service among tremendous amount available item vast growth information Internet well number visitor website add key challenge recommender system producing accurate...
[0.021156195551156998, -0.018920622766017914, -0.025686029344797134, 0.010952256619930267, -0.008208991959691048, -0.03683309629559517, 0.013347703963518143, 0.022213084623217583, 0.024835387244820595, 0.002295314334332943, -0.038629163056612015, 0.029846066609025, -0.0038689197972416878, 0.04660201072692871, -0.032441...
40,425
40,425
['Jérôme Kunegis', 'Ernesto W. De Luca', 'Sahin Albayrak']
1006.5367v1
We define and study the link prediction problem in bipartite networks, specializing general link prediction algorithms to the bipartite case. In a graph, a link prediction function of two vertices denotes the similarity or proximity of the vertices. Common link prediction functions for general graphs are defined using ...
The Link Prediction Problem in Bipartite Networks
2,010
http://arxiv.org/pdf/1006.5367v1
Title Link Prediction Problem Bipartite Networks Summary define study link prediction problem bipartite network specializing general link prediction algorithm bipartite case graph link prediction function two vertex denotes similarity proximity vertex Common link prediction function general graph defined using path len...
[-0.020887264981865883, -0.013763255439698696, -0.04404943436384201, 0.024788018316030502, -0.04572137072682381, -0.059555575251579285, 0.0003048788639716804, 0.014871177263557911, 0.06635553389787674, -0.0238523930311203, -0.012899250723421574, 0.006487246137112379, -0.03370117023587227, 0.02789970301091671, 0.0270546...
40,426
40,426
['Laurent El Ghaoui', 'Vivian Viallon', 'Tarek Rabbani']
1009.3515v2
We investigate fast methods that allow to quickly eliminate variables (features) in supervised learning problems involving a convex loss function and a $l_1$-norm penalty, leading to a potentially substantial reduction in the number of variables prior to running the supervised learning algorithm. The methods are not he...
Safe Feature Elimination in Sparse Supervised Learning
2,010
http://arxiv.org/pdf/1009.3515v2
Title Safe Feature Elimination Sparse Supervised Learning Summary investigate fast method allow quickly eliminate variable feature supervised learning problem involving convex loss function l1norm penalty leading potentially substantial reduction number variable prior running supervised learning algorithm method heuris...
[-0.004182683303952217, 0.019323615357279778, 0.013433349318802357, 0.026084961369633675, 0.0019009689567610621, 0.012187328189611435, 0.016926968470215797, 0.013389825820922852, -0.003949007485061884, -0.002247439231723547, 0.019117632880806923, 0.03507605567574501, -0.004972449969500303, 0.093934565782547, -0.0291411...
40,427
40,427
['Konstantin Biatov']
1009.4719v1
This paper describes an effective unsupervised speaker indexing approach. We suggest a two stage algorithm to speed-up the state-of-the-art algorithm based on the Bayesian Information Criterion (BIC). In the first stage of the merging process a computationally cheap method based on the vector quantization (VQ) is used....
A Fast Audio Clustering Using Vector Quantization and Second Order Statistics
2,010
http://arxiv.org/pdf/1009.4719v1
Title Fast Audio Clustering Using Vector Quantization Second Order Statistics Summary paper describes effective unsupervised speaker indexing approach suggest two stage algorithm speedup stateoftheart algorithm based Bayesian Information Criterion BIC first stage merging process computationally cheap method based vecto...
[-0.07355310022830963, 0.002461535856127739, -0.023719482123851776, 0.01576410047709942, 0.009297529235482216, 0.0028405864723026752, 0.05474359169602394, 0.016290586441755295, -0.026025623083114624, 0.01948051154613495, -0.028040455654263496, 0.001579692238010466, 0.030901042744517326, -0.035311292856931686, 0.0020179...
40,428
40,428
['S. M. Kamruzzaman', 'A. N. M. Rezaul Karim', 'Md. Saiful Islam', 'Md. Emdadul Haque']
1009.4972v1
Speech recognition and speaker identification are important for authentication and verification in security purpose, but they are difficult to achieve. Speaker identification methods can be divided into text-independent and text-dependent. This paper presents a technique of text-dependent speaker identification using M...
Speaker Identification using MFCC-Domain Support Vector Machine
2,010
http://arxiv.org/pdf/1009.4972v1
Title Speaker Identification using MFCCDomain Support Vector Machine Summary Speech recognition speaker identification important authentication verification security purpose difficult achieve Speaker identification method divided textindependent textdependent paper present technique textdependent speaker identification...
[-0.02607783116400242, -0.03939138352870941, 0.0004652688803616911, 0.059521060436964035, -0.013520976528525352, 0.0010644126450642943, 0.04558362439274788, -0.032348401844501495, -0.020375605672597885, -0.004435076843947172, 0.0039507425390183926, -0.010391661897301674, 0.0551946721971035, -0.015581436455249786, -0.03...
40,429
40,429
['Matthew D. Hoffman']
1009.5761v1
In certain applications it is useful to fit multinomial distributions to observed data with a penalty term that encourages sparsity. For example, in probabilistic latent audio source decomposition one may wish to encode the assumption that only a few latent sources are active at any given time. The standard heuristic o...
Approximate Maximum A Posteriori Inference with Entropic Priors
2,010
http://arxiv.org/pdf/1009.5761v1
Title Approximate Maximum Posteriori Inference Entropic Priors Summary certain application useful fit multinomial distribution observed data penalty term encourages sparsity example probabilistic latent audio source decomposition one may wish encode assumption latent source active given time standard heuristic applying...
[-0.028505655005574226, 0.05937838554382324, 0.030276861041784286, -0.008063510060310364, -0.008356306701898575, -0.01043867040425539, 0.002014602767303586, 0.027927594259381294, -0.10546406358480453, -0.0030313758179545403, -0.028950681909918785, 0.02949124574661255, 0.026805799454450607, -0.03663402795791626, 0.01931...
40,430
40,430
['Sudipto Guha', 'Kamesh Munagala', 'Martin Pal']
1011.1161v3
In this paper we initiate the study of optimization of bandit type problems in scenarios where the feedback of a play is not immediately known. This arises naturally in allocation problems which have been studied extensively in the literature, albeit in the absence of delays in the feedback. We study this problem in th...
Multiarmed Bandit Problems with Delayed Feedback
2,010
http://arxiv.org/pdf/1011.1161v3
Title Multiarmed Bandit Problems Delayed Feedback Summary paper initiate study optimization bandit type problem scenario feedback play immediately known arises naturally allocation problem studied extensively literature albeit absence delay feedback study problem Bayesian setting presence delay solution provable guaran...
[0.03368421643972397, 0.06928133219480515, -0.021639585494995117, -0.03520025312900543, -0.004692803602665663, -0.03717969357967377, 0.036099065095186234, 0.012210307642817497, -0.0208603348582983, -0.006836640182882547, 0.0051446715369820595, -0.006070876494050026, -0.05805251747369766, 0.0240118820220232, 0.016658492...
40,431
40,431
['Jacob Abernethy', 'Peter L. Bartlett', 'Elad Hazan']
1011.1936v1
We consider the celebrated Blackwell Approachability Theorem for two-player games with vector payoffs. We show that Blackwell's result is equivalent, via efficient reductions, to the existence of "no-regret" algorithms for Online Linear Optimization. Indeed, we show that any algorithm for one such problem can be effici...
Blackwell Approachability and Low-Regret Learning are Equivalent
2,010
http://arxiv.org/pdf/1011.1936v1
Title Blackwell Approachability LowRegret Learning Equivalent Summary consider celebrated Blackwell Approachability Theorem twoplayer game vector payoff show Blackwells result equivalent via efficient reduction existence noregret algorithm Online Linear Optimization Indeed show algorithm one problem efficiently convert...
[-0.025725586339831352, 0.05888979509472847, -0.03264343738555908, -0.01007288321852684, -0.026690054684877396, -0.043660104274749756, -0.006411109119653702, 0.0330372154712677, -0.02700350247323513, 0.04223168268799782, 0.004927211906760931, 0.0478442907333374, -0.012339211069047451, 0.019279396161437035, -0.013761328...
40,432
40,432
['Ofer Dekel', 'Ran Gilad-Bachrach', 'Ohad Shamir', 'Lin Xiao']
1012.1370v1
The standard model of online prediction deals with serial processing of inputs by a single processor. However, in large-scale online prediction problems, where inputs arrive at a high rate, an increasingly common necessity is to distribute the computation across several processors. A non-trivial challenge is to design ...
Robust Distributed Online Prediction
2,010
http://arxiv.org/pdf/1012.1370v1
Title Robust Distributed Online Prediction Summary standard model online prediction deal serial processing input single processor However largescale online prediction problem input arrive high rate increasingly common necessity distribute computation across several processor nontrivial challenge design distributed algo...
[-0.010637825354933739, 0.049490492790937424, 0.0008666878566145897, -0.026396922767162323, -0.011116912588477135, -0.02983156032860279, 0.04066409170627594, -0.006100289989262819, 0.005309219937771559, 0.017921870574355125, 0.020653370767831802, -0.02095170132815838, -0.013956001959741116, 0.06987880170345306, -0.0026...
40,433
40,433
['Jaydip Sen', 'P. Balamuralidhar', 'M. Girish Chandra', 'Harihara S. G.', 'Harish Reddy']
1012.2514v1
In a dynamic heterogeneous environment, such as pervasive and ubiquitous computing, context-aware adaptation is a key concept to meet the varying requirements of different users. Connectivity is an important context source that can be utilized for optimal management of diverse networking resources. Application QoS (Qua...
Context Aware End-to-End Connectivity Management
2,010
http://arxiv.org/pdf/1012.2514v1
Title Context Aware EndtoEnd Connectivity Management Summary dynamic heterogeneous environment pervasive ubiquitous computing contextaware adaptation key concept meet varying requirement different user Connectivity important context source utilized optimal management diverse networking resource Application QoS Quality ...
[-0.011681274510920048, -0.049678653478622437, -0.03071439638733864, -0.033424459397792816, -0.019076120108366013, -0.043289314955472946, 0.001731115160509944, -0.040650781244039536, -0.05092458426952362, -0.034483667463064194, 0.0021821721456944942, 0.04263055697083473, 0.009747988544404507, 0.004107656888663769, -0.0...
40,434
40,434
['Yuesheng Xu', 'Haizhang Zhang', 'Qinghui Zhang']
1102.1324v1
This paper studies the construction of a refinement kernel for a given operator-valued reproducing kernel such that the vector-valued reproducing kernel Hilbert space of the refinement kernel contains that of the given one as a subspace. The study is motivated from the need of updating the current operator-valued repro...
Refinement of Operator-valued Reproducing Kernels
2,011
http://arxiv.org/pdf/1102.1324v1
Title Refinement Operatorvalued Reproducing Kernels Summary paper study construction refinement kernel given operatorvalued reproducing kernel vectorvalued reproducing kernel Hilbert space refinement kernel contains given one subspace study motivated need updating current operatorvalued reproducing kernel multitask lea...
[-0.04058101773262024, 0.002062075072899461, -0.016445217654109, 0.02876720391213894, 0.0002765085082501173, -0.04929583892226219, 0.02022245153784752, 0.0357179120182991, -0.0421418733894825, 0.013021821156144142, -0.00048807289567776024, 0.018163098022341728, 0.02438286319375038, 0.010919943451881409, -0.000424589001...
40,435
40,435
['Cem Tekin', 'Mingyan Liu']
1102.3508v1
In this paper we study the online learning problem involving rested and restless multiarmed bandits with multiple plays. The system consists of a single player/user and a set of K finite-state discrete-time Markov chains (arms) with unknown state spaces and statistics. At each time step the player can play M arms. The ...
Online Learning of Rested and Restless Bandits
2,011
http://arxiv.org/pdf/1102.3508v1
Title Online Learning Rested Restless Bandits Summary paper study online learning problem involving rested restless multiarmed bandit multiple play system consists single playeruser set K finitestate discretetime Markov chain arm unknown state space statistic time step player play arm objective user decide step K arm p...
[-0.013716498389840126, 0.014653187245130539, -0.01814441569149494, -0.06391686946153641, -0.007744941394776106, -0.00031889244564808905, -0.0026929499581456184, 0.019389919936656952, -0.0374709852039814, -0.0056071835570037365, -0.0017937240190804005, 0.016076184809207916, -0.053290970623493195, 0.006135296076536179, ...
40,436
40,436
['Manas A. Pathak', 'Mehrbod Sharifi', 'Bhiksha Raj']
1102.4021v2
Email is a private medium of communication, and the inherent privacy constraints form a major obstacle in developing effective spam filtering methods which require access to a large amount of email data belonging to multiple users. To mitigate this problem, we envision a privacy preserving spam filtering system, where ...
Privacy Preserving Spam Filtering
2,011
http://arxiv.org/pdf/1102.4021v2
Title Privacy Preserving Spam Filtering Summary Email private medium communication inherent privacy constraint form major obstacle developing effective spam filtering method require access large amount email data belonging multiple user mitigate problem envision privacy preserving spam filtering system server able trai...
[0.05335427448153496, 0.043901629745960236, -0.027694644406437874, 0.0027716667391359806, -0.04806884005665779, -0.015176949091255665, 0.06387370824813843, 0.024198807775974274, 0.07549292594194412, -0.07344169914722443, 0.010013969615101814, 0.004214583430439234, -0.04396883398294449, 0.055237364023923874, -0.01993795...
40,437
40,437
['Vincent Gripon', 'Claude Berrou']
1102.4240v1
Coded recurrent neural networks with three levels of sparsity are introduced. The first level is related to the size of messages, much smaller than the number of available neurons. The second one is provided by a particular coding rule, acting as a local constraint in the neural activity. The third one is a characteris...
Sparse neural networks with large learning diversity
2,011
http://arxiv.org/pdf/1102.4240v1
Title Sparse neural network large learning diversity Summary Coded recurrent neural network three level sparsity introduced first level related size message much smaller number available neuron second one provided particular coding rule acting local constraint neural activity third one characteristic low final connecti...
[-0.013131286017596722, 0.017885690554976463, 0.004463877063244581, 0.042043741792440414, 0.004247666802257299, -0.023627111688256264, 0.02238982543349266, -0.004302415531128645, -0.026891011744737625, -0.01997786946594715, 0.014922210946679115, -0.024358754977583885, 0.014609604142606258, 0.08070508390665054, 0.025863...
40,438
40,438
['Arvind Narayanan', 'Elaine Shi', 'Benjamin I. P. Rubinstein']
1102.4374v1
This paper describes the winning entry to the IJCNN 2011 Social Network Challenge run by Kaggle.com. The goal of the contest was to promote research on real-world link prediction, and the dataset was a graph obtained by crawling the popular Flickr social photo sharing website, with user identities scrubbed. By de-anony...
Link Prediction by De-anonymization: How We Won the Kaggle Social Network Challenge
2,011
http://arxiv.org/pdf/1102.4374v1
Title Link Prediction Deanonymization Kaggle Social Network Challenge Summary paper describes winning entry IJCNN 2011 Social Network Challenge run Kagglecom goal contest promote research realworld link prediction dataset graph obtained crawling popular Flickr social photo sharing website user identity scrubbed deanony...
[0.04381046071648598, 0.029864363372325897, -0.029891015961766243, -0.0009283635881729424, -0.02701568230986595, -0.0576777458190918, 0.024023933336138725, -0.0024084579199552536, 0.05766779184341431, -0.0030509072821587324, -0.016661521047353745, 0.05632903799414635, -0.023327475413680077, 0.05920444801449776, 0.04541...
40,439
40,439
['Radim Řeh{ů}řek']
1102.5597v1
With the explosion of the size of digital dataset, the limiting factor for decomposition algorithms is the \emph{number of passes} over the input, as the input is often stored out-of-core or even off-site. Moreover, we're only interested in algorithms that operate in \emph{constant memory} w.r.t. to the input size, so ...
Fast and Faster: A Comparison of Two Streamed Matrix Decomposition Algorithms
2,011
http://arxiv.org/pdf/1102.5597v1
Title Fast Faster Comparison Two Streamed Matrix Decomposition Algorithms Summary explosion size digital dataset limiting factor decomposition algorithm emphnumber pass input input often stored outofcore even offsite Moreover interested algorithm operate emphconstant memory wrt input size arbitrarily large input proces...
[-0.017311837524175644, 0.044995538890361786, -0.019035859033465385, 0.014180190861225128, -0.017251810058951378, 0.017915785312652588, 0.014299463480710983, 0.02958986721932888, -0.02453421615064144, -0.03218042477965355, 0.008186760358512402, -0.0033274174202233553, 0.03849092498421669, 0.02254636213183403, -0.041635...
40,440
40,440
['Wahiba Ben Abdessalem Karaa']
1102.5728v1
This paper introduces a named entity recognition approach in textual corpus. This Named Entity (NE) can be a named: location, person, organization, date, time, etc., characterized by instances. A NE is found in texts accompanied by contexts: words that are left or right of the NE. The work mainly aims at identifying co...
Named Entity Recognition Using Web Document Corpus
2,011
http://arxiv.org/pdf/1102.5728v1
Title Named Entity Recognition Using Web Document Corpus Summary paper introduces named entity recognition approach textual corpus Named Entity NE named location person organization date time etc characterized instance NE found text accompanied context word left right NE work mainly aim identifying context inducing NEs...
[0.050518445670604706, 0.0003095581487286836, 0.018921896815299988, 0.0272502563893795, -0.001999913016334176, 0.06367451697587967, 0.007857482880353928, 0.0531846359372139, 0.0032762589398771524, -0.08147557824850082, -6.393183866748586e-05, -0.01582382805645466, 0.017247553914785385, -0.02485865168273449, -0.01239887...
40,441
40,441
['Jeff M. Phillips', 'Suresh Venkatasubramanian']
1103.1625v2
This document reviews the definition of the kernel distance, providing a gentle introduction tailored to a reader with background in theoretical computer science, but limited exposure to technology more common to machine learning, functional analysis and geometric measure theory. The key aspect of the kernel distance d...
A Gentle Introduction to the Kernel Distance
2,011
http://arxiv.org/pdf/1103.1625v2
Title Gentle Introduction Kernel Distance Summary document review definition kernel distance providing gentle introduction tailored reader background theoretical computer science limited exposure technology common machine learning functional analysis geometric measure theory key aspect kernel distance developed interpr...
[-0.03430366516113281, -0.03003401681780815, -0.05438952520489693, 0.04043497517704964, -0.06585151702165604, -0.03915788605809212, 0.014056208543479443, 0.02273094281554222, -0.02181917615234852, 0.01391929667443037, 0.08696269243955612, 0.013857759535312653, 0.03587630018591881, 0.028803743422031403, -0.0310881864279...
40,442
40,442
['Hua Ouyang', 'Alexander Gray']
1105.2274v1
In this paper, we focus on the question of the extent to which online learning can benefit from distributed computing. We focus on the setting in which $N$ agents online-learn cooperatively, where each agent only has access to its own data. We propose a generic data-distributed online learning meta-algorithm. We then i...
Data-Distributed Weighted Majority and Online Mirror Descent
2,011
http://arxiv.org/pdf/1105.2274v1
Title DataDistributed Weighted Majority Online Mirror Descent Summary paper focus question extent online learning benefit distributed computing focus setting N agent onlinelearn cooperatively agent access data propose generic datadistributed online learning metaalgorithm introduce Distributed Weighted Majority Distribu...
[-0.008095704019069672, 0.06287980079650879, -0.009347891435027122, -0.05815209075808525, -0.05492109805345535, -0.009143348783254623, 0.06093887984752655, -0.018909255042672157, -0.011472145095467567, 0.010743960738182068, 0.007794508710503578, 0.005796857178211212, 0.00866268016397953, -0.0007949781138449907, -0.0356...
40,443
40,443
['Nathan Keller', 'Elchanan Mossel', 'Tomer Schlank']
1105.2651v1
The entropy/influence conjecture, raised by Friedgut and Kalai in 1996, seeks to relate two different measures of concentration of the Fourier coefficients of a Boolean function. Roughly saying, it claims that if the Fourier spectrum is "smeared out", then the Fourier coefficients are concentrated on "high" levels. In ...
A Note on the Entropy/Influence Conjecture
2,011
http://arxiv.org/pdf/1105.2651v1
Title Note EntropyInfluence Conjecture Summary entropyinfluence conjecture raised Friedgut Kalai 1996 seek relate two different measure concentration Fourier coefficient Boolean function Roughly saying claim Fourier spectrum smeared Fourier coefficient concentrated high level note generalize conjecture biased product m...
[-0.046705618500709534, 0.00589234521612525, -0.022418469190597534, 0.023377349600195885, -0.058923956006765366, 0.0031981344800442457, -0.03570364788174629, -0.01303965412080288, -0.06407658010721207, -0.011812582612037659, -0.00047706643817946315, -0.018081791698932648, 0.032433804124593735, -0.04607446491718292, 0.0...
40,444
40,444
['Alejandro Chinea', 'Elka Korutcheva']
1106.1113v1
Graph-based representations of images have recently acquired an important role for classification purposes within the context of machine learning approaches. The underlying idea is to consider that relevant information of an image is implicitly encoded into the relationships between more basic entities that compose by ...
Complexity Analysis of Vario-eta through Structure
2,011
http://arxiv.org/pdf/1106.1113v1
Title Complexity Analysis Varioeta Structure Summary Graphbased representation image recently acquired important role classification purpose within context machine learning approach underlying idea consider relevant information image implicitly encoded relationship basic entity compose whole image classification proble...
[0.028029773384332657, -0.026818692684173584, -0.03118187189102173, 0.036237932741642, -0.009286381304264069, -0.013351447880268097, -0.005831494461745024, 0.004818853922188282, 0.017185529693961143, -0.027568114921450615, 0.013751523569226265, -0.040592897683382034, 0.039839111268520355, 0.07224604487419128, 0.0141162...
40,445
40,445
['Andrei N. Soklakov']
1106.2882v1
The recent crisis and the following flight to simplicity put most derivative businesses around the world under considerable pressure. We argue that the traditional modeling techniques must be extended to include product design. We propose a quantitative framework for creating products which meet the challenge of being ...
Learning, investments and derivatives
2,011
http://arxiv.org/pdf/1106.2882v1
Title Learning investment derivative Summary recent crisis following flight simplicity put derivative business around world considerable pressure argue traditional modeling technique must extended include product design propose quantitative framework creating product meet challenge optimal investor point view remaining...
[-0.012964962981641293, 0.0032524282578378916, -0.04890253767371178, -0.015313949435949326, 0.017314137890934944, -0.03863007202744484, 0.08063368499279022, -0.0018333394546061754, -0.04829415678977966, 0.015777284279465675, 0.03724389895796776, 0.022425612434744835, -0.02029533125460148, 0.1180749461054802, 0.01286941...
40,446
40,446
['Sławomir Staworko', 'Piotr Wieczorek']
1106.3725v3
We investigate the problem of learning XML queries, path queries and tree pattern queries, from examples given by the user. A learning algorithm takes on the input a set of XML documents with nodes annotated by the user and returns a query that selects the nodes in a manner consistent with the annotation. We study two ...
Learning XML Twig Queries
2,011
http://arxiv.org/pdf/1106.3725v3
Title Learning XML Twig Queries Summary investigate problem learning XML query path query tree pattern query example given user learning algorithm take input set XML document node annotated user return query selects node manner consistent annotation study two learning setting differ type annotation first setting user m...
[0.01732443831861019, -0.018138404935598373, -0.04111219942569733, 0.01595362462103367, -0.0036186419893056154, -0.01312639657407999, -0.03088042326271534, 0.04128484055399895, 0.020431367680430412, -0.020378833636641502, 0.0020067139994353056, 0.08088325709104538, -0.028626717627048492, 0.040031857788562775, -0.020722...
40,447
40,447
['Haizhang Zhang', 'Liang Zhao']
1106.4075v1
To help understand various reproducing kernels used in applied sciences, we investigate the inclusion relation of two reproducing kernel Hilbert spaces. Characterizations in terms of feature maps of the corresponding reproducing kernels are established. A full table of inclusion relations among widely-used translation ...
On the Inclusion Relation of Reproducing Kernel Hilbert Spaces
2,011
http://arxiv.org/pdf/1106.4075v1
Title Inclusion Relation Reproducing Kernel Hilbert Spaces Summary help understand various reproducing kernel used applied science investigate inclusion relation two reproducing kernel Hilbert space Characterizations term feature map corresponding reproducing kernel established full table inclusion relation among widel...
[-0.04156538471579552, -0.021605517715215683, -0.004156978335231543, 0.00632587680593133, -0.021801991388201714, -0.007710620760917664, -0.004362447187304497, 0.022654244676232338, 0.01049487479031086, 0.016329044476151466, -0.031856928020715714, 0.003924938850104809, 0.033610958606004715, 0.00492897117510438, 0.020578...
40,448
40,448
['Feng Niu', 'Benjamin Recht', 'Christopher Re', 'Stephen J. Wright']
1106.5730v2
Stochastic Gradient Descent (SGD) is a popular algorithm that can achieve state-of-the-art performance on a variety of machine learning tasks. Several researchers have recently proposed schemes to parallelize SGD, but all require performance-destroying memory locking and synchronization. This work aims to show using no...
HOGWILD!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
2,011
http://arxiv.org/pdf/1106.5730v2
Title HOGWILD LockFree Approach Parallelizing Stochastic Gradient Descent Summary Stochastic Gradient Descent SGD popular algorithm achieve stateoftheart performance variety machine learning task Several researcher recently proposed scheme parallelize SGD require performancedestroying memory locking synchronization wor...
[0.004885129630565643, 0.03907415643334389, 0.010512537322938442, 0.013273569755256176, -0.0060016317293047905, -0.007038085255771875, 0.04037092998623848, -0.018938684836030006, 0.00581173924729228, -0.006551637779921293, -0.002931873081251979, -0.0016088885022327304, 0.010615813545882702, -0.03279566392302513, -0.041...
40,449
40,449
['Jeff M. Phillips', 'Parasaran Raman', 'Suresh Venkatasubramanian']
1108.0017v1
We provide a new framework for generating multiple good quality partitions (clusterings) of a single data set. Our approach decomposes this problem into two components, generating many high-quality partitions, and then grouping these partitions to obtain k representatives. The decomposition makes the approach extremely...
Generating a Diverse Set of High-Quality Clusterings
2,011
http://arxiv.org/pdf/1108.0017v1
Title Generating Diverse Set HighQuality Clusterings Summary provide new framework generating multiple good quality partition clustering single data set approach decomposes problem two component generating many highquality partition grouping partition obtain k representative decomposition make approach extremely modula...
[-0.04027153551578522, 0.022773627191781998, -0.05941138416528702, -0.004752946086227894, -0.00579395005479455, -0.012312346138060093, 0.0947207659482956, 0.036089882254600525, 0.050908591598272324, -0.012774871662259102, 0.011398497968912125, 0.03726387023925781, 0.0017199504654854536, 0.053205907344818115, 0.01884794...
40,450
40,450
['Yao Wu', 'Qiang Yan', 'Danny Bickson', 'Yucheng Low', 'Qing Yang']
1108.2580v2
This paper describes the solution method taken by LeBuSiShu team for track1 in ACM KDD CUP 2011 contest (resulting in the 5th place). We identified two main challenges: the unique item taxonomy characteristics as well as the large data set size.To handle the item taxonomy, we present a novel method called Matrix Factor...
Efficient Multicore Collaborative Filtering
2,011
http://arxiv.org/pdf/1108.2580v2
Title Efficient Multicore Collaborative Filtering Summary paper describes solution method taken LeBuSiShu team track1 ACM KDD CUP 2011 contest resulting 5th place identified two main challenge unique item taxonomy characteristic well large data set sizeTo handle item taxonomy present novel method called Matrix Factoriz...
[-0.005671095103025436, -0.017328254878520966, -0.007633903529495001, 0.00183896126691252, -0.023127306252717972, -0.006768211256712675, 0.015174571424722672, 0.03263797610998154, 0.0047952583990991116, -0.008574698120355606, -0.05167130008339882, 0.034689418971538544, -0.019638169556856155, 0.03711261972784996, -0.062...
40,451
40,451
['Meir Perez', 'Tshilidzi Marwala']
1108.4545v1
The Fuzzy Gene Filter (FGF) is an optimised Fuzzy Inference System designed to rank genes in order of differential expression, based on expression data generated in a microarray experiment. This paper examines the effectiveness of the FGF for feature selection using various classification architectures. The FGF is comp...
The fuzzy gene filter: A classifier performance assesment
2,011
http://arxiv.org/pdf/1108.4545v1
Title fuzzy gene filter classifier performance assesment Summary Fuzzy Gene Filter FGF optimised Fuzzy Inference System designed rank gene order differential expression based expression data generated microarray experiment paper examines effectiveness FGF feature selection using various classification architecture FGF ...
[0.010194633156061172, 0.015802031382918358, -0.05856885015964508, -0.026595301926136017, 0.009030790999531746, 0.02499782107770443, 0.07045989483594894, 0.05544121563434601, 0.044243600219488144, -0.0030834400095045567, 0.04584519937634468, 0.009285980835556984, 0.02706695720553398, 0.0508841909468174, -0.026519143953...
40,452
40,452
['Mlungisi Duma', 'Bhekisipho Twala', 'Tshilidzi Marwala']
1108.4551v1
The Ripper algorithm is designed to generate rule sets for large datasets with many features. However, it was shown that the algorithm struggles with classification performance in the presence of missing data. The algorithm struggles to classify instances when the quality of the data deteriorates as a result of increas...
Improving the performance of the ripper in insurance risk classification : A comparitive study using feature selection
2,011
http://arxiv.org/pdf/1108.4551v1
Title Improving performance ripper insurance risk classification comparitive study using feature selection Summary Ripper algorithm designed generate rule set large datasets many feature However shown algorithm struggle classification performance presence missing data algorithm struggle classify instance quality data d...
[-0.03227360174059868, 0.03989262506365776, -0.04628752917051315, -0.005425445735454559, 0.0038489813450723886, 0.027814051136374474, 0.021330172196030617, 0.05551028251647949, -0.017009466886520386, -0.00890099536627531, 0.06992381066083908, 0.0352860689163208, 0.022255616262555122, 0.06659016758203506, -0.01257159281...
40,453
40,453
['Massimo Melucci']
1108.5784v1
The Probability Ranking Principle states that the document set with the highest values of probability of relevance optimizes information retrieval effectiveness given the probabilities are estimated as accurately as possible. The key point of the principle is the separation of the document set into two subsets with a g...
Probability Ranking in Vector Spaces
2,011
http://arxiv.org/pdf/1108.5784v1
Title Probability Ranking Vector Spaces Summary Probability Ranking Principle state document set highest value probability relevance optimizes information retrieval effectiveness given probability estimated accurately possible key point principle separation document set two subset given level fallout highest recall pap...
[0.03646327182650566, 0.022608602419495583, -0.024488143622875214, 0.010572039522230625, -0.021908624097704887, 0.006696054246276617, 0.03295242413878441, 0.014115915633738041, -0.024763092398643494, -0.051713407039642334, 0.027809174731373787, -0.006469155661761761, 0.003717937972396612, 0.010345447808504105, 0.002115...
40,454
40,454
['Dean Foster', 'Alexander Rakhlin']
1108.6088v1
We present an algorithm which attains O(\sqrt{T}) internal (and thus external) regret for finite games with partial monitoring under the local observability condition. Recently, this condition has been shown by (Bartok, Pal, and Szepesvari, 2011) to imply the O(\sqrt{T}) rate for partial monitoring games against an i.i...
No Internal Regret via Neighborhood Watch
2,011
http://arxiv.org/pdf/1108.6088v1
Title Internal Regret via Neighborhood Watch Summary present algorithm attains OsqrtT internal thus external regret finite game partial monitoring local observability condition Recently condition shown Bartok Pal Szepesvari 2011 imply OsqrtT rate partial monitoring game iid opponent author conjectured hold nonstochasti...
[-0.028885263949632645, 0.07380794733762741, -0.0077223023399710655, -0.03094932809472084, -0.058986663818359375, -0.02986287698149681, -0.03293204307556152, 0.0199408121407032, 0.005857918411493301, 0.024358395487070084, 0.024003084748983383, 0.023645352572202682, -0.06771283596754074, 0.007228503935039043, 0.00605176...
40,455
40,455
['Zenglin Xu', 'Feng Yan', 'Yuan', 'Qi']
1108.6296v2
Tensor decomposition is a powerful computational tool for multiway data analysis. Many popular tensor decomposition approaches---such as the Tucker decomposition and CANDECOMP/PARAFAC (CP)---amount to multi-linear factorization. They are insufficient to model (i) complex interactions between data entities, (ii) various...
Infinite Tucker Decomposition: Nonparametric Bayesian Models for Multiway Data Analysis
2,011
http://arxiv.org/pdf/1108.6296v2
Title Infinite Tucker Decomposition Nonparametric Bayesian Models Multiway Data Analysis Summary Tensor decomposition powerful computational tool multiway data analysis Many popular tensor decomposition approachessuch Tucker decomposition CANDECOMPPARAFAC CPamount multilinear factorization insufficient model complex in...
[-0.0397573858499527, 0.052809204906225204, -0.021833444014191628, -0.0011870188172906637, -0.0009496218990534544, 0.03206805884838104, 0.04665861278772354, 0.03684527426958084, -0.018377449363470078, 0.021731767803430557, 0.017620794475078583, 0.01850310154259205, -0.0054091508500278, 0.039467811584472656, -0.00530939...
40,456
40,456
['Matus Telgarsky']
1110.1514v1
This manuscript investigates the relationship between Blackwell Approachability, a stochastic vector-valued repeated game, and minimax theory, a single-play scalar-valued scenario. First, it is established in a general setting --- one not permitting invocation of minimax theory --- that Blackwell's Approachability Theo...
Blackwell Approachability and Minimax Theory
2,011
http://arxiv.org/pdf/1110.1514v1
Title Blackwell Approachability Minimax Theory Summary manuscript investigates relationship Blackwell Approachability stochastic vectorvalued repeated game minimax theory singleplay scalarvalued scenario First established general setting one permitting invocation minimax theory Blackwells Approachability Theorem genera...
[0.007566201500594616, -0.05488952621817589, -0.034085098654031754, 0.008108546026051044, -0.021745888516306877, -0.06042355298995972, -0.023612461984157562, 0.002361014485359192, -0.03276069462299347, 0.045144904404878616, -0.008495624177157879, 0.004608249291777611, -0.023165876045823097, 0.005720503628253937, 0.0049...
40,457
40,457
['G. Kesidis', 'A. Kurve']
1110.1781v1
We consider unsupervised crowdsourcing performance based on the model wherein the responses of end-users are essentially rated according to how their responses correlate with the majority of other responses to the same subtasks/questions. In one setting, we consider an independent sequence of identically distributed cr...
A Study of Unsupervised Adaptive Crowdsourcing
2,011
http://arxiv.org/pdf/1110.1781v1
Title Study Unsupervised Adaptive Crowdsourcing Summary consider unsupervised crowdsourcing performance based model wherein response endusers essentially rated according response correlate majority response subtasksquestions one setting consider independent sequence identically distributed crowdsourcing assignment meta...
[0.04418741539120674, 0.009188026189804077, -0.009902602061629295, -0.010587126016616821, -0.023117342963814735, -0.01513640396296978, -0.013867486268281937, -0.01949048973619938, -0.04021793231368065, -0.028012383729219437, -0.03026706725358963, 0.021158087998628616, -0.00368309672921896, 0.003373813582584262, -0.0059...
40,458
40,458
['Dip Narayan Ray', 'Somajyoti Majumder', 'Sumit Mukhopadhyay']
1110.1796v1
The use of mobile robots is being popular over the world mainly for autonomous explorations in hazardous/ toxic or unknown environments. This exploration will be more effective and efficient if the explorations in unknown environment can be aided with the learning from past experiences. Currently reinforcement learning...
A Behavior-based Approach for Multi-agent Q-learning for Autonomous Exploration
2,011
http://arxiv.org/pdf/1110.1796v1
Title Behaviorbased Approach Multiagent Qlearning Autonomous Exploration Summary use mobile robot popular world mainly autonomous exploration hazardous toxic unknown environment exploration effective efficient exploration unknown environment aided learning past experience Currently reinforcement learning getting accept...
[0.021604781970381737, -0.03661254793405533, -0.009501338005065918, -0.04273391142487526, 0.01019483245909214, -0.014268930070102215, 0.019427388906478882, -0.040378108620643616, -0.02276538498699665, -0.0389939621090889, -0.04908838868141174, 0.06408891081809998, -0.04603172466158867, 0.016105175018310547, -0.02001627...
40,459
40,459
['Ohad Shamir']
1110.2392v2
We provide a variant of Azuma's concentration inequality for martingales, in which the standard boundedness requirement is replaced by the milder requirement of a subgaussian tail.
A Variant of Azuma's Inequality for Martingales with Subgaussian Tails
2,011
http://arxiv.org/pdf/1110.2392v2
Title Variant Azumas Inequality Martingales Subgaussian Tails Summary provide variant Azumas concentration inequality martingale standard boundedness requirement replaced milder requirement subgaussian tail Authors 0 Ahmed Osman Wojciech Samek 1 Ji Young Lee Franck Dernoncourt 2 Iulian Vlad Serban Tim Klinger Gerald Te...
[-0.009964226745069027, -0.02787928096950054, -0.042300958186388016, -0.01910429075360298, 0.0029517330694943666, -0.0035435666795819998, -0.029368741437792778, -0.0008533383370377123, -0.00923298578709364, 0.061040546745061874, 0.062027379870414734, 0.002602418651804328, -0.04415658116340637, -0.016748568043112755, -0...
40,460
40,460
['Parul Agarwal', 'M. Afshar Alam', 'Ranjit Biswas']
1110.2610v1
Clustering is an unsupervised technique of Data Mining. It means grouping similar objects together and separating the dissimilar ones. Each object in the data set is assigned a class label in the clustering process using a distance measure. This paper has captured the problems that are faced in real when clustering alg...
Issues,Challenges and Tools of Clustering Algorithms
2,011
http://arxiv.org/pdf/1110.2610v1
Title IssuesChallenges Tools Clustering Algorithms Summary Clustering unsupervised technique Data Mining mean grouping similar object together separating dissimilar one object data set assigned class label clustering process using distance measure paper captured problem faced real clustering algorithm implemented also ...
[0.00741969607770443, 0.009911803528666496, -0.029504045844078064, 0.058433692902326584, -0.004695613868534565, -0.023943843320012093, 0.06507731229066849, 0.025506513193249702, -0.013212619349360466, -0.0013171392492949963, 0.02764797769486904, 0.053061287850141525, 0.045003343373537064, 0.006497976370155811, -0.04481...
40,461
40,461
['K. Usha Rani']
1110.2626v1
One of the important techniques of Data mining is Classification. Many real world problems in various fields such as business, science, industry and medicine can be solved by using classification approach. Neural Networks have emerged as an important tool for classification. The advantages of Neural Networks helps for ...
Analysis of Heart Diseases Dataset using Neural Network Approach
2,011
http://arxiv.org/pdf/1110.2626v1
Title Analysis Heart Diseases Dataset using Neural Network Approach Summary One important technique Data mining Classification Many real world problem various field business science industry medicine solved using classification approach Neural Networks emerged important tool classification advantage Neural Networks hel...
[-0.009211324155330658, 0.024765290319919586, -0.03653160482645035, 0.009831558912992477, 0.02508270926773548, 0.025086427107453346, 0.028974195942282677, 0.017673401162028313, 0.010109558701515198, 0.039405789226293564, 0.03780662268400192, 0.030956100672483444, 0.051442716270685196, 0.025810671970248222, -0.013694418...
40,462
40,462
['Daniel Hsu', 'Sham M. Kakade', 'Tong Zhang']
1110.2842v1
We prove an exponential probability tail inequality for positive semidefinite quadratic forms in a subgaussian random vector. The bound is analogous to one that holds when the vector has independent Gaussian entries.
A tail inequality for quadratic forms of subgaussian random vectors
2,011
http://arxiv.org/pdf/1110.2842v1
Title tail inequality quadratic form subgaussian random vector Summary prove exponential probability tail inequality positive semidefinite quadratic form subgaussian random vector bound analogous one hold vector independent Gaussian entry Authors 0 Ahmed Osman Wojciech Samek 1 Ji Young Lee Franck Dernoncourt 2 Iulian V...
[-0.013714881613850594, 0.0008419348741881549, -0.024875205010175705, 0.03204478323459625, -0.016443030908703804, 0.0143289091065526, -0.025625670328736305, -0.006597098428755999, -0.01632525771856308, -0.015618914738297462, 0.03477989137172699, -0.006898859981447458, -0.05313068628311157, 0.029160695150494576, -0.0024...
40,463
40,463
['Christos Boutsidis', 'Anastasios Zouzias', 'Michael W. Mahoney', 'Petros Drineas']
1110.2897v3
We study the topic of dimensionality reduction for $k$-means clustering. Dimensionality reduction encompasses the union of two approaches: \emph{feature selection} and \emph{feature extraction}. A feature selection based algorithm for $k$-means clustering selects a small subset of the input features and then applies $k...
Randomized Dimensionality Reduction for k-means Clustering
2,011
http://arxiv.org/pdf/1110.2897v3
Title Randomized Dimensionality Reduction kmeans Clustering Summary study topic dimensionality reduction kmeans clustering Dimensionality reduction encompasses union two approach emphfeature selection emphfeature extraction feature selection based algorithm kmeans clustering selects small subset input feature applies k...
[-0.036905281245708466, -0.036915361881256104, -0.05960271880030632, 0.029135333374142647, 0.02626664564013481, -0.02281659096479416, 0.05244864150881767, 0.018098406493663788, -0.0009620037162676454, 0.02119303122162819, 0.05181598290801048, 0.039168838411569595, 0.022225718945264816, 0.04023393988609314, -0.028229072...
40,464
40,464
['Peng Cheng']
1110.3001v1
We propose a first-order method for stochastic strongly convex optimization that attains $O(1/n)$ rate of convergence, analysis show that the proposed method is simple, easily to implement, and in worst case, asymptotically four times faster than its peers. We derive this method from several intuitive observations that...
Step size adaptation in first-order method for stochastic strongly convex programming
2,011
http://arxiv.org/pdf/1110.3001v1
Title Step size adaptation firstorder method stochastic strongly convex programming Summary propose firstorder method stochastic strongly convex optimization attains O1n rate convergence analysis show proposed method simple easily implement worst case asymptotically four time faster peer derive method several intuitive...
[-0.007402130868285894, 0.011893377639353275, 0.003955082967877388, -0.0035117112565785646, 0.007753078360110521, -0.01196298934519291, -0.02796700969338417, -0.00035457091871649027, -0.01854410581290722, 0.01386135071516037, 0.011233935132622719, 0.011598821729421616, -0.00897132232785225, 0.09838902950286865, -0.0072...
40,465
40,465
['C. Staiger', 'S. Cadot', 'R. Kooter', 'M. Dittrich', 'T. Mueller', 'G. W. Klau', 'L. F. A. Wessels']
1110.3717v2
Recently, several classifiers that combine primary tumor data, like gene expression data, and secondary data sources, such as protein-protein interaction networks, have been proposed for predicting outcome in breast cancer. In these approaches, new composite features are typically constructed by aggregating the express...
A critical evaluation of network and pathway based classifiers for outcome prediction in breast cancer
2,011
http://arxiv.org/pdf/1110.3717v2
Title critical evaluation network pathway based classifier outcome prediction breast cancer Summary Recently several classifier combine primary tumor data like gene expression data secondary data source proteinprotein interaction network proposed predicting outcome breast cancer approach new composite feature typically...
[0.006087668240070343, 0.06257890164852142, -0.03029298409819603, -0.06149847432971001, -0.009559647180140018, 0.019785230979323387, 0.053882647305727005, 0.010427337139844894, -0.00768444687128067, 0.023588895797729492, 0.038237668573856354, -0.03276168182492256, 0.013506680727005005, 0.090369813144207, -0.02916058339...
40,466
40,466
['Wouter Lueks', 'Bassam Mokbel', 'Michael Biehl', 'Barbara Hammer']
1110.3917v1
The growing number of dimensionality reduction methods available for data visualization has recently inspired the development of quality assessment measures, in order to evaluate the resulting low-dimensional representation independently from a methods' inherent criteria. Several (existing) quality measures can be (re)...
How to Evaluate Dimensionality Reduction? - Improving the Co-ranking Matrix
2,011
http://arxiv.org/pdf/1110.3917v1
Title Evaluate Dimensionality Reduction Improving Coranking Matrix Summary growing number dimensionality reduction method available data visualization recently inspired development quality assessment measure order evaluate resulting lowdimensional representation independently method inherent criterion Several existing ...
[-0.05682862922549248, -0.015106872655451298, -0.030593179166316986, 0.032980695366859436, -0.009262307547032833, -0.006148871034383774, 0.06567328423261642, 0.03204140067100525, -0.009414350613951683, 0.02742868848145008, -0.015505214221775532, 0.0197173859924078, 0.04507710039615631, 0.035846419632434845, -0.00662798...
40,467
40,467
['Georgios C. Chasparis', 'Ari Arapostathis', 'Jeff S. Shamma']
1110.4412v1
We consider the problem of distributed convergence to efficient outcomes in coordination games through dynamics based on aspiration learning. Under aspiration learning, a player continues to play an action as long as the rewards received exceed a specified aspiration level. Here, the aspiration level is a fading memory...
Aspiration Learning in Coordination Games
2,011
http://arxiv.org/pdf/1110.4412v1
Title Aspiration Learning Coordination Games Summary consider problem distributed convergence efficient outcome coordination game dynamic based aspiration learning aspiration learning player continues play action long reward received exceed specified aspiration level aspiration level fading memory average past reward l...
[0.0005067908787168562, 0.025769149884581566, -0.015248771756887436, -0.03698379173874855, -0.03039039857685566, -0.025763902813196182, -0.04938212409615517, -0.01985963061451912, -0.0334230400621891, 0.02858392708003521, -0.000359184603439644, 0.033625341951847076, -0.020903032273054123, 0.021530935540795326, -0.02582...
40,468
40,468
['Sébastien Bubeck', 'Damien Ernst', 'Aurélien Garivier']
1110.5447v1
We consider an original problem that arises from the issue of security analysis of a power system and that we name optimal discovery with probabilistic expert advice. We address it with an algorithm based on the optimistic paradigm and the Good-Turing missing mass estimator. We show that this strategy uniformly attains...
Optimal discovery with probabilistic expert advice
2,011
http://arxiv.org/pdf/1110.5447v1
Title Optimal discovery probabilistic expert advice Summary consider original problem arises issue security analysis power system name optimal discovery probabilistic expert advice address algorithm based optimistic paradigm GoodTuring missing mass estimator show strategy uniformly attains optimal discovery rate macros...
[-0.0012803737772628665, 0.04294147342443466, -0.00937445554882288, -0.007802258245646954, -0.014699519611895084, -0.0787624940276146, 0.01362521480768919, -0.026773786172270775, -0.025512948632240295, 0.03810250386595726, 0.09093555808067322, 0.01744815520942211, -0.028730474412441254, 0.050946786999702454, 0.01390909...
40,469
40,469
['Sanjeev Arora', 'Rong Ge', 'Ravi Kannan', 'Ankur Moitra']
1111.0952v1
In the Nonnegative Matrix Factorization (NMF) problem we are given an $n \times m$ nonnegative matrix $M$ and an integer $r > 0$. Our goal is to express $M$ as $A W$ where $A$ and $W$ are nonnegative matrices of size $n \times r$ and $r \times m$ respectively. In some applications, it makes sense to ask instead for the...
Computing a Nonnegative Matrix Factorization -- Provably
2,011
http://arxiv.org/pdf/1111.0952v1
Title Computing Nonnegative Matrix Factorization Provably Summary Nonnegative Matrix Factorization NMF problem given n time nonnegative matrix integer r 0 goal express W W nonnegative matrix size n time r r time respectively application make sense ask instead product AW approximate ie approximately minimize normM AWF n...
[-0.08133860677480698, 0.02188250981271267, 0.0048426054418087006, 0.03408069908618927, -0.02228177711367607, -0.027285335585474968, -0.01738961972296238, 0.0029941389802843332, -0.029346348717808723, 0.005450672935694456, -0.00837821327149868, 0.009839549660682678, -0.01139398105442524, 0.04886729270219803, -0.0035888...
40,470
40,470
['Lisa Hellerstein', 'Devorah Kletenik', 'Linda Sellie', 'Rocco Servedio']
1111.1124v1
We prove a new structural lemma for partial Boolean functions $f$, which we call the seed lemma for DNF. Using the lemma, we give the first subexponential algorithm for proper learning of DNF in Angluin's Equivalence Query (EQ) model. The algorithm has time and query complexity $2^{(\tilde{O}{\sqrt{n}})}$, which is opt...
Tight Bounds on Proper Equivalence Query Learning of DNF
2,011
http://arxiv.org/pdf/1111.1124v1
Title Tight Bounds Proper Equivalence Query Learning DNF Summary prove new structural lemma partial Boolean function f call seed lemma DNF Using lemma give first subexponential algorithm proper learning DNF Angluins Equivalence Query EQ model algorithm time query complexity 2tildeOsqrtn optimal also give new result cer...
[-0.021013574674725533, 0.05939974635839462, -0.024251198396086693, -0.0017215233528986573, -0.03751815855503082, -0.017360946163535118, -0.002020846586674452, 0.025274382904171944, 0.007154291495680809, -0.013926906511187553, 0.025458354502916336, 0.06036722660064697, -0.013162296265363693, 0.06681164354085922, -0.007...
40,471
40,471
['Yuyang Wang', 'Roni Khardon', 'Pavlos Protopapas']
1111.1315v2
Many real world problems exhibit patterns that have periodic behavior. For example, in astrophysics, periodic variable stars play a pivotal role in understanding our universe. An important step when analyzing data from such processes is the problem of identifying the period: estimating the period of a periodic function...
Nonparametric Bayesian Estimation of Periodic Functions
2,011
http://arxiv.org/pdf/1111.1315v2
Title Nonparametric Bayesian Estimation Periodic Functions Summary Many real world problem exhibit pattern periodic behavior example astrophysics periodic variable star play pivotal role understanding universe important step analyzing data process problem identifying period estimating period periodic function based noi...
[-0.020405437797307968, 0.044719673693180084, -0.021982142701745033, 0.022782498970627785, 0.012237870134413242, -0.0130709707736969, 0.022722819820046425, 0.035347580909729004, -0.0914686769247055, 0.05058387294411659, 0.019915280863642693, -0.005873583722859621, 0.030791016295552254, 0.03179618716239929, -0.021457318...
40,472
40,472
['Shipra Agrawal', 'Navin Goyal']
1111.1797v3
The multi-armed bandit problem is a popular model for studying exploration/exploitation trade-off in sequential decision problems. Many algorithms are now available for this well-studied problem. One of the earliest algorithms, given by W. R. Thompson, dates back to 1933. This algorithm, referred to as Thompson Samplin...
Analysis of Thompson Sampling for the multi-armed bandit problem
2,011
http://arxiv.org/pdf/1111.1797v3
Title Analysis Thompson Sampling multiarmed bandit problem Summary multiarmed bandit problem popular model studying explorationexploitation tradeoff sequential decision problem Many algorithm available wellstudied problem One earliest algorithm given W R Thompson date back 1933 algorithm referred Thompson Sampling natu...
[0.007474137935787439, 0.02580147422850132, -0.01992236077785492, -0.05787891894578934, 0.0033037387765944004, -0.024110887199640274, -0.014801020734012127, 0.01576211117208004, -0.00411974499002099, -0.022831320762634277, -0.01940758526325226, 0.027255989611148834, -0.04420897737145424, -0.015092381276190281, 0.011187...
40,473
40,473
['Sanmay Das', 'Allen Lavoie', 'Malik Magdon-Ismail']
1111.2092v1
As a major source for information on virtually any topic, Wikipedia serves an important role in public dissemination and consumption of knowledge. As a result, it presents tremendous potential for people to promulgate their own points of view; such efforts may be more subtle than typical vandalism. In this paper, we in...
Pushing Your Point of View: Behavioral Measures of Manipulation in Wikipedia
2,011
http://arxiv.org/pdf/1111.2092v1
Title Pushing Point View Behavioral Measures Manipulation Wikipedia Summary major source information virtually topic Wikipedia serf important role public dissemination consumption knowledge result present tremendous potential people promulgate point view effort may subtle typical vandalism paper introduce new behaviora...
[0.0835721492767334, 0.055036064237356186, -0.002005439018830657, -0.006926395930349827, -0.01183839701116085, -0.022137992084026337, 0.015819290652871132, -0.01462238933891058, -0.020081324502825737, -0.03359947353601456, 0.05719330906867981, 0.023280905559659004, -0.012879989109933376, 0.0589054636657238, -0.02030234...
40,474
40,474
['Song Liu', 'Peter Flach', 'Nello Cristianini']
1111.2111v2
In this paper we introduce a generic model for multiplicative algorithms which is suitable for the MapReduce parallel programming paradigm. We implement three typical machine learning algorithms to demonstrate how similarity comparison, gradient descent, power method and other classic learning techniques fit this model...
Generic Multiplicative Methods for Implementing Machine Learning Algorithms on MapReduce
2,011
http://arxiv.org/pdf/1111.2111v2
Title Generic Multiplicative Methods Implementing Machine Learning Algorithms MapReduce Summary paper introduce generic model multiplicative algorithm suitable MapReduce parallel programming paradigm implement three typical machine learning algorithm demonstrate similarity comparison gradient descent power method class...
[-0.0046873618848621845, 0.027820125222206116, -0.01428820751607418, -0.024819057434797287, -0.04636220633983612, 0.02041865885257721, 0.03429890796542168, 0.022974791005253792, -0.026705268770456314, -0.025074755772948265, 0.03262483701109886, -0.002980815712362528, 0.017595242708921432, -0.008348620496690273, 0.00209...
40,475
40,475
['Rong Jin', 'Tianbao Yang', 'Mehrdad Mahdavi', 'Yu-Feng Li', 'Zhi-Hua Zhou']
1111.2262v4
We develop two approaches for analyzing the approximation error bound for the Nystr\"{o}m method, one based on the concentration inequality of integral operator, and one based on the compressive sensing theory. We show that the approximation error, measured in the spectral norm, can be improved from $O(N/\sqrt{m})$ to ...
Improved Bound for the Nystrom's Method and its Application to Kernel Classification
2,011
http://arxiv.org/pdf/1111.2262v4
Title Improved Bound Nystroms Method Application Kernel Classification Summary develop two approach analyzing approximation error bound Nystrom method one based concentration inequality integral operator one based compressive sensing theory show approximation error measured spectral norm improved ONsqrtm ONm1 rho case ...
[-0.004557634238153696, -0.021713348105549812, -0.03685520961880684, 0.03963432461023331, 0.034161582589149475, -0.026348356157541275, -0.02824351377785206, 0.05017203837633133, 0.0051629734225571156, 0.04059290885925293, 0.0036451066844165325, 0.014908810146152973, 0.050394147634506226, -0.06617944687604904, -0.049263...
40,476
40,476
['Jacob Abernethy', 'Rafael M. Frongillo']
1111.2664v1
Machine Learning competitions such as the Netflix Prize have proven reasonably successful as a method of "crowdsourcing" prediction tasks. But these competitions have a number of weaknesses, particularly in the incentive structure they create for the participants. We propose a new approach, called a Crowdsourced Learni...
A Collaborative Mechanism for Crowdsourcing Prediction Problems
2,011
http://arxiv.org/pdf/1111.2664v1
Title Collaborative Mechanism Crowdsourcing Prediction Problems Summary Machine Learning competition Netflix Prize proven reasonably successful method crowdsourcing prediction task competition number weakness particularly incentive structure create participant propose new approach called Crowdsourced Learning Mechanism...
[0.043289538472890854, 0.0932631865143776, -0.028352640569210052, -0.0008137922268360853, -0.038014695048332214, -0.037009093910455704, 0.03944040462374687, 0.03198350965976715, -0.030697038397192955, -0.025985516607761383, 0.004614692181348801, 0.045650187879800797, -0.0033454925287514925, 0.06984755396842957, 0.00437...
40,477
40,477
['Marcos A. Domingues', 'Alipio Mario Jorge', 'Carlos Soares']
1111.2948v2
Traditionally, recommender systems for the Web deal with applications that have two dimensions, users and items. Based on access logs that relate these dimensions, a recommendation model can be built and used to identify a set of N items that will be of interest to a certain user. In this paper we propose a method to c...
Using Contextual Information as Virtual Items on Top-N Recommender Systems
2,011
http://arxiv.org/pdf/1111.2948v2
Title Using Contextual Information Virtual Items TopN Recommender Systems Summary Traditionally recommender system Web deal application two dimension user item Based access log relate dimension recommendation model built used identify set N item interest certain user paper propose method complement information access l...
[0.0010286709293723106, -0.018283464014530182, -0.012580196373164654, 0.01220325194299221, -0.0020166137255728245, -0.019559934735298157, 0.07376115024089813, 0.0025184904225170612, -9.555765427649021e-05, -0.03892987594008446, -0.055147312581539154, 0.01899905502796173, -0.004921970423310995, 0.07588223367929459, -0.0...
40,478
40,478
['Matthew D. Hoffman', 'Andrew Gelman']
1111.4246v1
Hamiltonian Monte Carlo (HMC) is a Markov chain Monte Carlo (MCMC) algorithm that avoids the random walk behavior and sensitivity to correlated parameters that plague many MCMC methods by taking a series of steps informed by first-order gradient information. These features allow it to converge to high-dimensional targe...
The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo
2,011
http://arxiv.org/pdf/1111.4246v1
Title NoUTurn Sampler Adaptively Setting Path Lengths Hamiltonian Monte Carlo Summary Hamiltonian Monte Carlo HMC Markov chain Monte Carlo MCMC algorithm avoids random walk behavior sensitivity correlated parameter plague many MCMC method taking series step informed firstorder gradient information feature allow converg...
[-0.04490730166435242, -0.017258888110518456, 0.0009946783538907766, -0.04158544912934303, -0.008480435237288475, -0.059921395033597946, 0.02442995458841324, -0.0035952189937233925, -0.0454888753592968, 0.012552128173410892, 0.06712477654218674, 0.08803404122591019, 0.014949130825698376, 0.007245113141834736, 0.0088000...
40,479
40,479
['Francis Bach']
1111.6453v2
Submodular functions are relevant to machine learning for at least two reasons: (1) some problems may be expressed directly as the optimization of submodular functions and (2) the lovasz extension of submodular functions provides a useful set of regularization functions for supervised and unsupervised learning. In this...
Learning with Submodular Functions: A Convex Optimization Perspective
2,011
http://arxiv.org/pdf/1111.6453v2
Title Learning Submodular Functions Convex Optimization Perspective Summary Submodular function relevant machine learning least two reason 1 problem may expressed directly optimization submodular function 2 lovasz extension submodular function provides useful set regularization function supervised unsupervised learning...
[-0.03289031237363815, 0.013274774886667728, -0.03151078149676323, 0.01190526969730854, -0.013063324615359306, -0.03605254366993904, 0.0652390867471695, 0.03747071325778961, -0.005954527296125889, -0.025248924270272255, -7.607141742482781e-05, 0.010013964958488941, 0.018591832369565964, 0.01616767980158329, 0.018294421...
40,480
40,480
['Raajay Viswanathan', 'Prateek Jain', 'Srivatsan Laxman', 'Arvind Arasu']
1111.7295v2
In this paper, we consider the problem of estimating self-tuning histograms using query workloads. To this end, we propose a general learning theoretic formulation. Specifically, we use query feedback from a workload as training data to estimate a histogram with a small memory footprint that minimizes the expected erro...
A Learning Framework for Self-Tuning Histograms
2,011
http://arxiv.org/pdf/1111.7295v2
Title Learning Framework SelfTuning Histograms Summary paper consider problem estimating selftuning histogram using query workload end propose general learning theoretic formulation Specifically use query feedback workload training data estimate histogram small memory footprint minimizes expected error future query for...
[-0.060072608292102814, 0.00954117625951767, -0.018759265542030334, -0.00530221126973629, -0.005847308784723282, -0.016385110095143318, 0.04299316927790642, 0.016327660530805588, 0.04959607869386673, 0.05925130099058151, -0.004683283623307943, 0.03687116503715515, -0.040968190878629684, 0.030345970764756203, -0.0075696...
40,481
40,481
['Fabian Pedregosa', 'Gaël Varoquaux', 'Alexandre Gramfort', 'Vincent Michel', 'Bertrand Thirion', 'Olivier Grisel', 'Mathieu Blondel', 'Gilles Louppe', 'Peter Prettenhofer', 'Ron Weiss', 'Vincent Dubourg', 'Jake Vanderplas', 'Alexandre Passos', 'David Cournapeau', 'Matthieu Brucher', 'Matthieu Perrot', 'Édouard Duches...
1201.0490v3
Scikit-learn is a Python module integrating a wide range of state-of-the-art machine learning algorithms for medium-scale supervised and unsupervised problems. This package focuses on bringing machine learning to non-specialists using a general-purpose high-level language. Emphasis is put on ease of use, performance, d...
Scikit-learn: Machine Learning in Python
2,012
http://arxiv.org/pdf/1201.0490v3
Title Scikitlearn Machine Learning Python Summary Scikitlearn Python module integrating wide range stateoftheart machine learning algorithm mediumscale supervised unsupervised problem package focus bringing machine learning nonspecialists using generalpurpose highlevel language Emphasis put ease use performance documen...
[-0.023782163858413696, -0.019803276285529137, -0.007959798909723759, -0.011520509608089924, 0.006755550857633352, 0.0013267978793010116, 0.045777782797813416, 0.022119978442788124, 0.036771342158317566, -0.030418872833251953, 0.02885068580508232, 0.03819590061903, 0.017087725922465324, 0.09669168293476105, -0.00578111...
40,482
40,482
['Sung-eok Jeon', 'Chuanyi Ji']
1201.2575v2
For a large multi-hop wireless network, nodes are preferable to make distributed and localized link-scheduling decisions with only interactions among a small number of neighbors. However, for a slowly decaying channel and densely populated interferers, a small size neighborhood often results in nontrivial link outages ...
Joint Approximation of Information and Distributed Link-Scheduling Decisions in Wireless Networks
2,012
http://arxiv.org/pdf/1201.2575v2
Title Joint Approximation Information Distributed LinkScheduling Decisions Wireless Networks Summary large multihop wireless network node preferable make distributed localized linkscheduling decision interaction among small number neighbor However slowly decaying channel densely populated interferers small size neighbo...
[-0.03622050955891609, -0.023025421425700188, -0.012278659269213676, -0.03993459790945053, -0.03050478734076023, -0.03134356066584587, 0.033758778125047684, -0.030161961913108826, -0.005623499862849712, -0.008753341622650623, -0.005552194081246853, 0.04719796031713486, -0.003879763185977936, -0.030918510630726814, -0.0...
40,483
40,483
['Geetha Manjunatha', 'M Narasimha Murty', 'Dinkar Sitaram']
1201.2925v2
Most enterprise data is distributed in multiple relational databases with expert-designed schema. Using traditional single-table machine learning techniques over such data not only incur a computational penalty for converting to a 'flat' form (mega-join), even the human-specified semantic information present in the rel...
Combining Heterogeneous Classifiers for Relational Databases
2,012
http://arxiv.org/pdf/1201.2925v2
Title Combining Heterogeneous Classifiers Relational Databases Summary enterprise data distributed multiple relational database expertdesigned schema Using traditional singletable machine learning technique data incur computational penalty converting flat form megajoin even humanspecified semantic information present r...
[-0.0019899792969226837, 0.01491063367575407, -0.004826454911381006, -0.011555309407413006, 0.005849676672369242, 0.043990496546030045, 0.013232766650617123, 0.004107834305614233, -0.011569815687835217, -0.055312033742666245, -0.016532178968191147, -0.0019815561827272177, 0.00527906883507967, 0.015601527877151966, -0.0...
40,484
40,484
['Keqin Liu', 'Qing Zhao']
1201.4906v1
We consider the adaptive shortest-path routing problem in wireless networks under unknown and stochastically varying link states. In this problem, we aim to optimize the quality of communication between a source and a destination through adaptive path selection. Due to the randomness and uncertainties in the network dy...
Adaptive Shortest-Path Routing under Unknown and Stochastically Varying Link States
2,012
http://arxiv.org/pdf/1201.4906v1
Title Adaptive ShortestPath Routing Unknown Stochastically Varying Link States Summary consider adaptive shortestpath routing problem wireless network unknown stochastically varying link state problem aim optimize quality communication source destination adaptive path selection Due randomness uncertainty network dynami...
[-0.017567873001098633, -0.00883940514177084, -0.0022445530630648136, -0.09108451008796692, -0.02464054711163044, -0.07272612303495407, -0.03142378106713295, -0.0438542403280735, -0.0033798597287386656, -0.009286333806812763, 0.01535288617014885, 0.039543796330690384, -0.002583449939265847, -0.025008201599121094, 0.026...
40,485
40,485
['Bo Zhao', 'Benjamin I. P. Rubinstein', 'Jim Gemmell', 'Jiawei Han']
1203.0058v1
In practical data integration systems, it is common for the data sources being integrated to provide conflicting information about the same entity. Consequently, a major challenge for data integration is to derive the most complete and accurate integrated records from diverse and sometimes conflicting sources. We term ...
A Bayesian Approach to Discovering Truth from Conflicting Sources for Data Integration
2,012
http://arxiv.org/pdf/1203.0058v1
Title Bayesian Approach Discovering Truth Conflicting Sources Data Integration Summary practical data integration system common data source integrated provide conflicting information entity Consequently major challenge data integration derive complete accurate integrated record diverse sometimes conflicting source term...
[0.010385017842054367, 0.08955126255750656, -0.018067579716444016, 0.01762313023209572, -0.07671617716550827, 0.0025337664410471916, -0.0023278426378965378, 0.0038639591075479984, -0.014699184335768223, -0.015661459416151047, 0.06227758154273033, 0.042597174644470215, 0.06489681452512741, 0.06181884929537773, -0.054552...
40,486
40,486
['T Soni Madhulatha']
1203.2002v1
Clustering is a common technique for statistical data analysis, Clustering is the process of grouping the data into classes or clusters so that objects within a cluster have high similarity in comparison to one another, but are very dissimilar to objects in other clusters. Dissimilarities are assessed based on the attr...
Graph partitioning advance clustering technique
2,012
http://arxiv.org/pdf/1203.2002v1
Title Graph partitioning advance clustering technique Summary Clustering common technique statistical data analysis Clustering process grouping data class cluster object within cluster high similarity comparison one another dissimilar object cluster Dissimilarities assessed based attribute value describing object Often...
[-0.03013598546385765, -0.03673066943883896, -0.04833868145942688, 0.07500111311674118, -0.005695030093193054, -0.03104904107749462, 0.08227904886007309, 0.03879469260573387, 0.018108583986759186, -0.010135019198060036, 0.008239655755460262, 0.03880618140101433, 0.009322159923613071, 0.015360410325229168, -0.0137226730...
40,487
40,487
['Surjeet Kumar Yadav', 'Brijesh Bharadwaj', 'Saurabh Pal']
1203.2987v1
The main objective of higher education is to provide quality education to students. One way to achieve highest level of quality in higher education system is by discovering knowledge for prediction regarding enrolment of students in a course. This paper presents a data mining project to generate predictive models for s...
Mining Education Data to Predict Student's Retention: A comparative Study
2,012
http://arxiv.org/pdf/1203.2987v1
Title Mining Education Data Predict Students Retention comparative Study Summary main objective higher education provide quality education student One way achieve highest level quality higher education system discovering knowledge prediction regarding enrolment student course paper present data mining project generate ...
[0.04067019000649452, -0.014662496745586395, -0.04953932762145996, -0.01762125827372074, 0.00537005253136158, 0.025555957108736038, -0.00847665686160326, -0.02127707004547119, 0.04650519788265228, -0.03643127903342247, 0.06167490407824516, 0.042757999151945114, 0.013764122501015663, 0.06032932549715042, -0.037990868091...
40,488
40,488
['Hussein Saad', 'Amr Mohamed', 'Tamer ElBatt']
1203.3935v1
In this paper, we propose a distributed reinforcement learning (RL) technique called distributed power control using Q-learning (DPC-Q) to manage the interference caused by the femtocells on macro-users in the downlink. The DPC-Q leverages Q-Learning to identify the sub-optimal pattern of power allocation, which strive...
Distributed Cooperative Q-learning for Power Allocation in Cognitive Femtocell Networks
2,012
http://arxiv.org/pdf/1203.3935v1
Title Distributed Cooperative Qlearning Power Allocation Cognitive Femtocell Networks Summary paper propose distributed reinforcement learning RL technique called distributed power control using Qlearning DPCQ manage interference caused femtocells macrousers downlink DPCQ leverage QLearning identify suboptimal pattern ...
[-0.008876192383468151, 0.003917040769010782, -0.02563466690480709, -0.051042042672634125, -0.02085263654589653, -0.02443547733128071, 0.027497487142682076, -0.027508916333317757, -0.03425954282283783, 0.02560112252831459, -0.04615703225135803, 0.013955884613096714, -0.003834603587165475, -0.03349985182285309, -0.01880...
40,489
40,489
['Rajiv Khanna', 'Liang Zhang', 'Deepak Agarwal', 'Beechung Chen']
1203.5124v1
Predicting user affinity to items is an important problem in applications like content optimization, computational advertising, and many more. While bilinear random effect models (matrix factorization) provide state-of-the-art performance when minimizing RMSE through a Gaussian response model on explicit ratings data, ...
Parallel Matrix Factorization for Binary Response
2,012
http://arxiv.org/pdf/1203.5124v1
Title Parallel Matrix Factorization Binary Response Summary Predicting user affinity item important problem application like content optimization computational advertising many bilinear random effect model matrix factorization provide stateoftheart performance minimizing RMSE Gaussian response model explicit rating dat...
[0.016388026997447014, 0.002745093312114477, -0.041088495403528214, -0.006879848428070545, 0.019185442477464676, 0.028349382802844048, 0.02986254170536995, 0.029029332101345062, 0.03333953768014908, -0.043502893298864365, -0.06168684735894203, -0.022908983752131462, -0.025021854788064957, 0.05306939035654068, -0.025397...
40,490
40,490
['Philippe Lauret', 'Auline Rodler', 'Marc Muselli', 'Mathieu David', 'Hadja Diagne', 'Cyril Voyant']
1203.5446v1
This paper proposes to use a rather new modelling approach in the realm of solar radiation forecasting. In this work, two forecasting models: Autoregressive Moving Average (ARMA) and Neural Network (NN) models are combined to form a model committee. The Bayesian inference is used to affect a probability to each model i...
A Bayesian Model Committee Approach to Forecasting Global Solar Radiation
2,012
http://arxiv.org/pdf/1203.5446v1
Title Bayesian Model Committee Approach Forecasting Global Solar Radiation Summary paper proposes use rather new modelling approach realm solar radiation forecasting work two forecasting model Autoregressive Moving Average ARMA Neural Network NN model combined form model committee Bayesian inference used affect probabi...
[0.02921825461089611, 0.02299194410443306, -0.00038427949766628444, -0.04153311625123024, -0.011121361516416073, 0.021429704502224922, 0.031844835728406906, -0.02691463939845562, 0.057028718292713165, 0.008513273671269417, 0.029924195259809494, 0.060781676322221756, 0.055157966911792755, 0.019630858674645424, 0.0230339...
40,491
40,491
['Elad Hazan', 'Satyen Kale', 'Shai Shalev-Shwartz']
1204.0136v1
In several online prediction problems of recent interest the comparison class is composed of matrices with bounded entries. For example, in the online max-cut problem, the comparison class is matrices which represent cuts of a given graph and in online gambling the comparison class is matrices which represent permutati...
Near-Optimal Algorithms for Online Matrix Prediction
2,012
http://arxiv.org/pdf/1204.0136v1
Title NearOptimal Algorithms Online Matrix Prediction Summary several online prediction problem recent interest comparison class composed matrix bounded entry example online maxcut problem comparison class matrix represent cut given graph online gambling comparison class matrix represent permutation n team Another impo...
[-0.0083452258259058, 0.026189863681793213, -0.019127516075968742, -0.03993656486272812, -0.029751183465123177, -0.004520578775554895, 0.02971576526761055, 0.061239466071128845, 0.006080298218876123, 0.009584516286849976, -0.003417797153815627, -0.009761142544448376, -0.016216784715652466, 0.1017545685172081, -0.003915...
40,492
40,492
['Jia Zeng', 'Zhi-Qiang Liu', 'Xiao-Qin Cao']
1204.0170v2
Latent Dirichlet allocation (LDA) is a widely-used probabilistic topic modeling paradigm, and recently finds many applications in computer vision and computational biology. In this paper, we propose a fast and accurate batch algorithm, active belief propagation (ABP), for training LDA. Usually batch LDA algorithms requ...
A New Approach to Speeding Up Topic Modeling
2,012
http://arxiv.org/pdf/1204.0170v2
Title New Approach Speeding Topic Modeling Summary Latent Dirichlet allocation LDA widelyused probabilistic topic modeling paradigm recently find many application computer vision computational biology paper propose fast accurate batch algorithm active belief propagation ABP training LDA Usually batch LDA algorithm requ...
[0.045277900993824005, 0.01594267413020134, -0.0077589647844433784, 0.005606784950941801, -0.024153636768460274, 0.015051943250000477, 0.002328617963939905, 0.034713223576545715, -0.031079573556780815, -0.05897429212927818, 0.015014669857919216, -0.0005522504216060042, 0.001299244468100369, 0.046202536672353745, -0.006...
40,493
40,493
['Ali Soltan Mohammadi', 'L. Asadzadeh', 'D. D. Rezaee']
1204.1467v1
The rough-set theory proposed by Pawlak, has been widely used in dealing with data classification problems. The original rough-set model is, however, quite sensitive to noisy data. Tzung thus proposed deals with the problem of producing a set of fuzzy certain and fuzzy possible rules from quantitative data with a prede...
Learning Fuzzy β-Certain and β-Possible rules from incomplete quantitative data by rough sets
2,012
http://arxiv.org/pdf/1204.1467v1
Title Learning Fuzzy βCertain βPossible rule incomplete quantitative data rough set Summary roughset theory proposed Pawlak widely used dealing data classification problem original roughset model however quite sensitive noisy data Tzung thus proposed deal problem producing set fuzzy certain fuzzy possible rule quantita...
[-0.008604594506323338, 0.04992089793086052, -0.03573612868785858, 0.005504218395799398, 0.0520445816218853, 0.00623143557459116, -0.02128736302256584, 0.0543900802731514, -0.060818158090114594, -0.016900120303034782, 0.055998582392930984, -0.04552559554576874, 0.03377960994839668, 0.032802995294332504, -0.024321582168...
40,494
40,494
['Paulo F. C. Tilles', 'Jose F. Fontanari']
1204.1564v4
An explanation for the acquisition of word-object mappings is the associative learning in a cross-situational scenario. Here we present analytical results of the performance of a simple associative learning algorithm for acquiring a one-to-one mapping between $N$ objects and $N$ words based solely on the co-occurrence ...
Minimal model of associative learning for cross-situational lexicon acquisition
2,012
http://arxiv.org/pdf/1204.1564v4
Title Minimal model associative learning crosssituational lexicon acquisition Summary explanation acquisition wordobject mapping associative learning crosssituational scenario present analytical result performance simple associative learning algorithm acquiring onetoone mapping N object N word based solely cooccurrence...
[0.028988344594836235, 0.006461146287620068, -0.0003566641826182604, 0.08050159364938736, -0.033397313207387924, 0.016601983457803726, 0.0038417575415223837, 0.020140618085861206, 0.003987549804151058, -0.048212360590696335, -0.03716030344367027, -0.034772176295518875, 0.033422909677028656, 0.012743574567139149, -0.005...
40,495
40,495
['Shalini Puri', 'Sona Kaushik']
1204.2058v1
In this new and current era of technology, advancements and techniques, efficient and effective text document classification is becoming a challenging and highly required area to capably categorize text documents into mutually exclusive categories. Fuzzy similarity provides a way to find the similarity of features amon...
A technical study and analysis on fuzzy similarity based models for text classification
2,012
http://arxiv.org/pdf/1204.2058v1
Title technical study analysis fuzzy similarity based model text classification Summary new current era technology advancement technique efficient effective text document classification becoming challenging highly required area capably categorize text document mutually exclusive category Fuzzy similarity provides way f...
[0.04494453966617584, -0.014866065233945847, -0.031974103301763535, 0.01804627664387226, -0.039590466767549515, 0.03703761473298073, 0.009517819620668888, 0.05883296951651573, 0.017060106620192528, -0.10225218534469604, -0.00307719805277884, 0.006457677576690912, 0.01104025263339281, -0.027271069586277008, -0.038901373...
40,496
40,496
['Shalini Puri']
1204.2061v1
Text Classification is a challenging and a red hot field in the current scenario and has great importance in text categorization applications. A lot of research work has been done in this field but there is a need to categorize a collection of text documents into mutually exclusive categories by extracting the concepts...
A Fuzzy Similarity Based Concept Mining Model for Text Classification
2,012
http://arxiv.org/pdf/1204.2061v1
Title Fuzzy Similarity Based Concept Mining Model Text Classification Summary Text Classification challenging red hot field current scenario great importance text categorization application lot research work done field need categorize collection text document mutually exclusive category extracting concept feature using...
[0.04370729625225067, 0.002435473958030343, -0.04265874996781349, 0.03567979484796524, -0.02158309519290924, 0.02992284670472145, -0.007792997173964977, 0.0372229665517807, -0.06161671504378319, -0.08113507181406021, -0.0008127085166051984, -0.0007746033370494843, 0.02052106335759163, 0.009924427606165409, -0.033610865...
40,497
40,497
['Valentina Fedorova', 'Alex Gammerman', 'Ilia Nouretdinov', 'Vladimir Vovk']
1204.3251v2
A standard assumption in machine learning is the exchangeability of data, which is equivalent to assuming that the examples are generated from the same probability distribution independently. This paper is devoted to testing the assumption of exchangeability on-line: the examples arrive one by one, and after receiving ...
Plug-in martingales for testing exchangeability on-line
2,012
http://arxiv.org/pdf/1204.3251v2
Title Plugin martingale testing exchangeability online Summary standard assumption machine learning exchangeability data equivalent assuming example generated probability distribution independently paper devoted testing assumption exchangeability online example arrive one one receiving example would like valid measure ...
[0.006676100194454193, 0.04274365305900574, -0.013664867728948593, 0.012571236118674278, -0.035303745418787, -0.007819239050149918, -0.00829676166176796, -0.009020675905048847, 0.01355015765875578, 0.014451977796852589, 0.06789512187242508, 0.028345778584480286, 0.019585028290748596, 0.09383845329284668, -0.03941883891...
40,498
40,498
['Maria-Florina Balcan', 'Avrim Blum', 'Shai Fine', 'Yishay Mansour']
1204.3514v3
We consider the problem of PAC-learning from distributed data and analyze fundamental communication complexity questions involved. We provide general upper and lower bounds on the amount of communication needed to learn well, showing that in addition to VC-dimension and covering number, quantities such as the teaching-...
Distributed Learning, Communication Complexity and Privacy
2,012
http://arxiv.org/pdf/1204.3514v3
Title Distributed Learning Communication Complexity Privacy Summary consider problem PAClearning distributed data analyze fundamental communication complexity question involved provide general upper lower bound amount communication needed learn well showing addition VCdimension covering number quantity teachingdimensio...
[-0.00014148751506581903, 0.04299638420343399, -0.013660618104040623, 0.02393127977848053, -0.040874503552913666, -0.022397814318537712, 0.060144875198602676, -0.05679234489798546, 0.006550255697220564, 0.006440427154302597, -0.02484811097383499, 0.020564937964081764, 0.010421697050333023, 0.030369697138667107, -0.0047...
40,499
40,499
['Seyda Ertekin', 'Haym Hirsh', 'Cynthia Rudin']
1204.3611v1
The problem of "approximating the crowd" is that of estimating the crowd's majority opinion by querying only a subset of it. Algorithms that approximate the crowd can intelligently stretch a limited budget for a crowdsourcing task. We present an algorithm, "CrowdSense," that works in an online fashion to dynamically sa...
Learning to Predict the Wisdom of Crowds
2,012
http://arxiv.org/pdf/1204.3611v1
Title Learning Predict Wisdom Crowds Summary problem approximating crowd estimating crowd majority opinion querying subset Algorithms approximate crowd intelligently stretch limited budget crowdsourcing task present algorithm CrowdSense work online fashion dynamically sample subset labelers based explorationexploitatio...
[0.05981838330626488, 0.0718386247754097, -0.02103772573173046, 0.00718695530667901, 0.00020797421166207641, -0.03053440898656845, -0.03037877194583416, 0.015119343996047974, -0.03035658784210682, -0.014003682881593704, 0.0013191935140639544, 0.013439824804663658, 0.006744399666786194, 0.05514524132013321, 0.0158067196...