title
string
paper_url
string
authors
list
type
string
primary_area
string
abstract
large_string
keywords
list
TL;DR
large_string
submission_number
int64
arxiv_id
string
arxiv_id_source
string
embedding
list
A Discriminative Latent Variable Model for Online Clustering
https://proceedings.mlr.press/v32/samdani14.html
[ "Rajhans Samdani", "Kai-Wei Chang", "Dan Roth" ]
null
null
This paper presents a latent variable structured prediction model for discriminative supervised clustering of items called the Latent Left-linking Model (L3M). We present an online clustering algorithm for L3M based on a feature-based item similarity function. We provide a learning framework for estimating the similari...
[]
null
1
null
null
[ 0.02436882257461548, -0.017543260008096695, -0.006952633615583181, 0.015739062801003456, 0.042216189205646515, 0.03834263235330582, -0.009240769781172276, 0.004330829717218876, -0.00473228981718421, -0.0010509949643164873, -0.01022334210574627, 0.009138601832091808, -0.056980375200510025, ...
Kernel Mean Estimation and Stein Effect
https://proceedings.mlr.press/v32/muandet14.html
[ "Krikamol Muandet", "Kenji Fukumizu", "Bharath Sriperumbudur", "Arthur Gretton", "Bernhard Schoelkopf" ]
null
null
A mean function in reproducing kernel Hilbert space (RKHS), or a kernel mean, is an important part of many algorithms ranging from kernel principal component analysis to Hilbert-space embedding of distributions. Given a finite sample, an empirical average is the standard estimate for the true kernel mean. We show that ...
[]
null
2
null
null
[ -0.046318419277668, -0.0020217287819832563, 0.020295610651373863, 0.014953752979636192, 0.037266749888658524, 0.036837391555309296, 0.053159814327955246, -0.009486976079642773, -0.020162049680948257, -0.040243469178676605, -0.019574923440814018, -0.01485881395637989, -0.06499124318361282, ...
Demystifying Information-Theoretic Clustering
https://proceedings.mlr.press/v32/steeg14.html
[ "Greg Ver Steeg", "Aram Galstyan", "Fei Sha", "Simon DeDeo" ]
null
null
We propose a novel method for clustering data which is grounded in information-theoretic principles and requires no parametric assumptions. Previous attempts to use information theory to define clusters in an assumption-free way are based on maximizing mutual information between data and cluster labels. We demonstrate...
[]
null
3
1310.4210
title_snapshot
[ -0.02262169122695923, 0.0015338932862505317, -0.006505098659545183, 0.05754297599196434, 0.05820578709244728, 0.044585440307855606, 0.0071479445323348045, -0.00685817701742053, -0.0296610277146101, -0.030829954892396927, -0.021498851478099823, 0.01215219683945179, -0.05292385444045067, 0.0...
Covering Number for Efficient Heuristic-based POMDP Planning
https://proceedings.mlr.press/v32/zhanga14.html
[ "Zongzhang Zhang", "David Hsu", "Wee Sun Lee" ]
null
null
The difficulty of POMDP planning depends on the size of the search space involved. Heuristics are often used to reduce the search space size and improve computational efficiency; however, there are few theoretical bounds on their effectiveness. In this paper, we use the covering number to characterize the size of the ...
[]
null
4
null
null
[ -0.0850820541381836, -0.0026173684746026993, -0.03730921447277069, 0.05293199419975281, 0.048862822353839874, 0.05511672422289848, 0.0005292503628879786, -0.01240041945129633, -0.028056852519512177, -0.058643799275159836, -0.022747976705431938, -0.02418682537972927, -0.06650175899267197, -...
The Coherent Loss Function for Classification
https://proceedings.mlr.press/v32/yanga14.html
[ "Wenzhuo Yang", "Melvyn Sim", "Huan Xu" ]
null
null
A prediction rule in binary classification that aims to achieve the lowest probability of misclassification involves minimizing over a non-convex, 0-1 loss function, which is typically a computationally intractable optimization problem. To address the intractability, previous methods consider minimizing the cumulative ...
[]
null
5
null
null
[ -0.028514279052615166, -0.018298182636499405, 0.005057649686932564, 0.03486321493983269, 0.024834292009472847, 0.028476690873503685, 0.018056558445096016, -0.027988890185952187, -0.033514734357595444, -0.0315607450902462, -0.029013661667704582, 0.008912000805139542, -0.0669446811079979, 0....
Fast Stochastic Alternating Direction Method of Multipliers
https://proceedings.mlr.press/v32/zhong14.html
[ "Wenliang Zhong", "James Kwok" ]
null
null
We propose a new stochastic alternating direction method of multipliers (ADMM) algorithm, which incrementally approximates the full gradient in the linearized ADMM formulation. Besides having a low per-iteration complexity as existing stochastic ADMM algorithms, it improves the convergence rate on convex problems from...
[]
null
6
1308.3558
title_snapshot
[ -0.03491176292300224, -0.013810597360134125, 0.014639235101640224, -0.009834405034780502, 0.009931867010891438, 0.07322084903717041, 0.02918277308344841, 0.005537377670407295, -0.051432639360427856, -0.04498077929019928, -0.003024633973836899, -0.02889414317905903, -0.043505121022462845, -...
Active Detection via Adaptive Submodularity
https://proceedings.mlr.press/v32/chena14.html
[ "Yuxin Chen", "Hiroaki Shioi", "Cesar Fuentes Montesinos", "Lian Pin Koh", "Serge Wich", "Andreas Krause" ]
null
null
Efficient detection of multiple object instances is one of the fundamental challenges in computer vision. For certain object categories, even the best automatic systems are yet unable to produce high-quality detection results, and fully manual annotation would be an expensive process. How can detection algorithms inter...
[]
null
7
null
null
[ 0.0038205792661756277, -0.022942636162042618, 0.0018579522147774696, 0.015315593220293522, 0.03513019159436226, 0.005123646464198828, 0.020348481833934784, -0.00497029535472393, -0.04053942486643791, -0.05037448927760124, -0.0535394512116909, 0.0026534441858530045, -0.0798768550157547, -0....
Accelerated Proximal Stochastic Dual Coordinate Ascent for Regularized Loss Minimization
https://proceedings.mlr.press/v32/shalev-shwartz14.html
[ "Shai Shalev-Shwartz", "Tong Zhang" ]
null
null
We introduce a proximal version of the stochastic dual coordinate ascent method and show how to accelerate the method using an inner-outer iteration procedure. We analyze the runtime of the framework and obtain rates that improve state-of-the-art results for various key machine learning optimization problems including ...
[]
null
8
1309.2375
title_snapshot
[ -0.014739345759153366, -0.016945071518421173, 0.025695567950606346, 0.042021315544843674, 0.04076981171965599, 0.043828871101140976, 0.014849124476313591, -0.01669585146009922, -0.01750916801393032, -0.040420908480882645, -0.011449122801423073, -0.008462940342724323, -0.037956152111291885, ...
An Adaptive Accelerated Proximal Gradient Method and its Homotopy Continuation for Sparse Optimization
https://proceedings.mlr.press/v32/lin14.html
[ "Qihang Lin", "Lin Xiao" ]
null
null
We first propose an adaptive accelerated proximal gradient(APG) method for minimizing strongly convex composite functions with unknown convexity parameters. This method incorporates a restarting scheme to automatically estimate the strong convexity parameter and achieves a nearly optimal iteration complexity. Then we c...
[]
null
9
null
null
[ -0.045880917459726334, -0.01718510314822197, 0.05340992286801338, 0.014694258570671082, 0.037124909460544586, 0.03695855662226677, 0.024446619674563408, 0.006234657019376755, -0.05560578778386116, -0.05621960759162903, -0.02456359937787056, -0.00848864670842886, -0.055453214794397354, -0.0...
Recurrent Convolutional Neural Networks for Scene Labeling
https://proceedings.mlr.press/v32/pinheiro14.html
[ "Pedro Pinheiro", "Ronan Collobert" ]
null
null
The goal of the scene labeling task is to assign a class label to each pixel in an image. To ensure a good visual coherence and a high class accuracy, it is essential for a model to capture long range pixel) label dependencies in images. In a feed-forward architecture, this can be achieved simply by considering a suf...
[]
null
10
null
null
[ 0.012783684767782688, -0.013212406076490879, -0.014304730109870434, 0.04185662418603897, 0.033679064363241196, 0.04126020520925522, -0.000661841535475105, 0.030361704528331757, -0.027129802852869034, -0.016217008233070374, -0.044058993458747864, -0.008443154394626617, -0.06360460817813873, ...
A Statistical Perspective on Algorithmic Leveraging
https://proceedings.mlr.press/v32/ma14.html
[ "Ping Ma", "Michael Mahoney", "Bin Yu" ]
null
null
One popular method for dealing with large-scale data sets is sampling. Using the empirical statistical leverage scores as an importance sampling distribution, the method of algorithmic leveraging samples and rescales rows/columns of data matrices to reduce the data size before performing computations on the subproblem....
[]
null
11
1306.5362
title_snapshot
[ -0.0027327905409038067, -0.03572462499141693, -0.01706513948738575, 0.019677430391311646, 0.057978466153144836, 0.017840474843978882, 0.01231385301798582, -0.014694924466311932, -0.008504260331392288, -0.04595091566443443, 0.007182316854596138, -0.04233187064528465, -0.07863681763410568, -...
Thompson Sampling for Complex Online Problems
https://proceedings.mlr.press/v32/gopalan14.html
[ "Aditya Gopalan", "Shie Mannor", "Yishay Mansour" ]
null
null
We consider stochastic multi-armed bandit problems with complex actions over a set of basic arms, where the decision maker plays a complex action rather than a basic arm in each round. The reward of the complex action is some function of the basic arms’ rewards, and the feedback observed may not necessarily be the rewa...
[]
null
12
null
null
[ -0.01266509760171175, 0.004511216655373573, -0.004786692094057798, 0.04712296277284622, 0.03990370035171509, 0.039815712720155716, 0.031002109870314598, -0.005420976784080267, -0.026239920407533646, -0.039918288588523865, -0.011453426443040371, 0.0068855504505336285, -0.062177155166864395, ...
Boosting multi-step autoregressive forecasts
https://proceedings.mlr.press/v32/taieb14.html
[ "Souhaib Ben Taieb", "Rob Hyndman" ]
null
null
Multi-step forecasts can be produced recursively by iterating a one-step model, or directly using a specific model for each horizon. Choosing between these two strategies is not an easy task since it involves a trade-off between bias and estimation variance over the forecast horizon. Using a nonlinear machine learning ...
[]
null
13
null
null
[ -0.0210976991802454, -0.03822632133960724, 0.01957538165152073, 0.007526834029704332, 0.05771385878324509, 0.05526750907301903, 0.05051451921463013, 0.007113004568964243, -0.0379432737827301, -0.06525129079818726, -0.003725288901478052, 0.0193988885730505, -0.07323738932609558, -0.00104582...
A Statistical Convergence Perspective of Algorithms for Rank Aggregation from Pairwise Data
https://proceedings.mlr.press/v32/rajkumar14.html
[ "Arun Rajkumar", "Shivani Agarwal" ]
null
null
There has been much interest recently in the problem of rank aggregation from pairwise data. A natural question that arises is: under what sorts of statistical assumptions do various rank aggregation algorithms converge to an ‘optimal’ ranking? In this paper, we consider this question in a natural setting where pairwis...
[]
null
14
null
null
[ -0.03859003260731697, -0.03197331726551056, 0.0365508571267128, 0.016977855935692787, 0.0030060068238526583, -0.017021991312503815, 0.0401301234960556, 0.0007607759907841682, -0.05430809408426285, -0.018732614815235138, -0.017210520803928375, -0.016299409791827202, -0.07340600341558456, -0...
Scaling Up Approximate Value Iteration with Options: Better Policies with Fewer Iterations
https://proceedings.mlr.press/v32/mann14.html
[ "Timothy Mann", "Shie Mannor" ]
null
null
We show how options, a class of control structures encompassing primitive and temporally extended actions, can play a valuable role in planning in MDPs with continuous state-spaces. Analyzing the convergence rate of Approximate Value Iteration with options reveals that for pessimistic initial value function estimates, ...
[]
null
15
null
null
[ -0.07485771179199219, -0.028933649882674217, -0.04051966965198517, 0.06638046354055405, 0.0757860466837883, 0.03893472999334335, 0.004872322082519531, -0.002531489124521613, -0.029605327174067497, -0.06535093486309052, -0.024417277425527573, -0.0012102966429665685, -0.07612577825784683, 0....
Latent Bandits.
https://proceedings.mlr.press/v32/maillard14.html
[ "Odalric-Ambrym Maillard", "Shie Mannor" ]
null
null
We consider a multi-armed bandit problem where the reward distributions are indexed by two sets –one for arms, one for type– and can be partitioned into a small number of clusters according to the type. First, we consider the setting where all reward distributions are known and all types have the same underlying cluste...
[]
null
16
null
null
[ -0.004674611613154411, -0.023204002529382706, 0.0006334254867397249, 0.06308045238256454, 0.04419531300663948, 0.04339715093374252, 0.006833449937403202, 0.004291007295250893, -0.027804862707853317, -0.0490165539085865, -0.01730523444712162, 0.009294402785599232, -0.03219298645853996, -0.0...
Fast Allocation of Gaussian Process Experts
https://proceedings.mlr.press/v32/nguyena14.html
[ "Trung Nguyen", "Edwin Bonilla" ]
null
null
We propose a scalable nonparametric Bayesian regression model based on a mixture of Gaussian process (GP) experts and the inducing points formalism underpinning sparse GP approximations. Each expert is augmented with a set of inducing points, and the allocation of data points to experts is defined probabilistically ba...
[]
null
17
null
null
[ -0.010526071302592754, -0.010184706188738346, -0.01410627644509077, 0.025437530130147934, 0.01740429177880287, 0.06313610821962357, 0.01029137335717678, -0.02311268076300621, -0.008155944757163525, -0.030586030334234238, -0.02918122336268425, 0.034645937383174896, -0.07067173719406128, 0.0...
Von Mises-Fisher Clustering Models
https://proceedings.mlr.press/v32/gopal14.html
[ "Siddharth Gopal", "Yiming Yang" ]
null
null
This paper proposes a suite of models for clustering high-dimensional data on a unit sphere based on Von Mises-Fisher (vMF) distribution and for discovering more intuitive clusters than existing approaches. The proposed models include a) A Bayesian formulation of vMF mixture that enables information sharing among clus...
[]
null
18
null
null
[ 0.006674848031252623, -0.005706466268748045, 0.025039933621883392, 0.04893667995929718, 0.038866061717271805, 0.06949491053819656, 0.031797025352716446, -0.0005543481092900038, -0.04921194911003113, -0.04334896430373192, -0.005577542819082737, -0.004335682839155197, -0.04620100557804108, 0...
Convergence rates for persistence diagram estimation in Topological Data Analysis
https://proceedings.mlr.press/v32/chazal14.html
[ "Frédéric Chazal", "Marc Glisse", "Catherine Labruère", "Bertrand Michel" ]
null
null
Computational topology has recently seen an important development toward data analysis, giving birth to Topological Data Analysis. Persistent homology appears as a fundamental tool in this field. We show that the use of persistent homology can be naturally considered in general statistical frameworks. We establish co...
[]
null
19
null
null
[ -0.029548373073339462, -0.0021140179596841335, -0.0061823101714253426, 0.024240897968411446, 0.037755049765110016, 0.02153673581779003, 0.008946498855948448, 0.02176683209836483, -0.014690042473375797, -0.03701449930667877, -0.00850469060242176, -0.007276100106537342, -0.08975224196910858, ...
Buffer k-d Trees: Processing Massive Nearest Neighbor Queries on GPUs
https://proceedings.mlr.press/v32/gieseke14.html
[ "Fabian Gieseke", "Justin Heinermann", "Cosmin Oancea", "Christian Igel" ]
null
null
We present a new approach for combining k-d trees and graphics processing units for nearest neighbor search. It is well known that a direct combination of these tools leads to a non-satisfying performance due to conditional computations and suboptimal memory accesses. To alleviate these problems, we propose a variant o...
[]
null
20
null
null
[ -0.043611664324998856, -0.011376332491636276, 0.006021418143063784, 0.03610583394765854, 0.016214121133089066, 0.02834029495716095, -0.010707201436161995, 0.03865044564008713, -0.012246145866811275, -0.0675957053899765, -0.03618060052394867, -0.008350472897291183, -0.051666464656591415, 0....
Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget
https://proceedings.mlr.press/v32/korattikara14.html
[ "Anoop Korattikara", "Yutian Chen", "Max Welling" ]
null
null
Can we make Bayesian posterior MCMC sampling more efficient when faced with very large datasets? We argue that computing the likelihood for N datapoints in the Metropolis-Hastings (MH) test to reach a single binary decision is computationally inefficient. We introduce an approximate MH rule based on a sequential hypoth...
[]
null
21
1304.5299
title_snapshot
[ -0.0004809324746020138, -0.0034848307259380817, -0.032860852777957916, 0.05077647045254707, 0.037604477256536484, 0.02696244604885578, 0.03602411970496178, -0.0037358927074819803, -0.03084290400147438, -0.06119436025619507, 0.017837196588516235, 0.009482001885771751, -0.0734865739941597, 0...
Understanding the Limiting Factors of Topic Modeling via Posterior Contraction Analysis
https://proceedings.mlr.press/v32/tang14.html
[ "Jian Tang", "Zhaoshi Meng", "Xuanlong Nguyen", "Qiaozhu Mei", "Ming Zhang" ]
null
null
Topic models such as the latent Dirichlet allocation (LDA) have become a standard staple in the modeling toolbox of machine learning. They have been applied to a vast variety of data sets, contexts, and tasks to varying degrees of success. However, to date there is almost no formal theory explicating the LDA’s behavior...
[]
null
22
null
null
[ 0.016052503138780594, -0.027484823018312454, -0.035998083651065826, 0.04079856351017952, 0.03611001744866371, 0.020571393892169, 0.018654605373740196, 0.003074079053476453, 0.010673553682863712, -0.011087271384894848, -0.027383318170905113, 0.011900638230144978, -0.041956719011068344, 0.01...
The Inverse Regression Topic Model
https://proceedings.mlr.press/v32/rabinovich14.html
[ "Maxim Rabinovich", "David Blei" ]
null
null
\citettaddy13mnir proposed multinomial inverse regression (MNIR) as a new model of annotated text based on the influence of metadata and response variables on the distribution of words in a document. While effective, MNIR has no way to exploit structure in the corpus to improve its predictions or facilitate exploratory...
[]
null
23
null
null
[ -0.012521536089479923, -0.05824431777000427, -0.024267537519335747, 0.0488891564309597, 0.03306080028414726, 0.004600097890943289, 0.02053980529308319, 0.0021454745437949896, -0.01862448640167713, -0.002201055409386754, -0.019921207800507545, 0.04597902297973633, -0.05659281089901924, 0.00...
A Consistent Histogram Estimator for Exchangeable Graph Models
https://proceedings.mlr.press/v32/chan14.html
[ "Stanley Chan", "Edoardo Airoldi" ]
null
null
Exchangeable graph models (ExGM) subsume a number of popular network models. The mathematical object that characterizes an ExGM is termed a graphon. Finding scalable estimators of graphons, provably consistent, remains an open issue. In this paper, we propose a histogram estimator of a graphon that is provably consist...
[]
null
24
1402.1888
title_snapshot
[ -0.006671191658824682, -0.016283772885799408, 0.002142473356798291, 0.026934167370200157, 0.023319561034440994, 0.017694154754281044, 0.008023226633667946, 0.03453719988465309, -0.05912346392869949, -0.10189038515090942, 0.042262524366378784, -0.023312903940677643, -0.06622793525457382, 0....
Latent Variable Copula Inference for Bundle Pricing from Retail Transaction Data
https://proceedings.mlr.press/v32/letham14.html
[ "Benjamin Letham", "Wei Sun", "Anshul Sheopuri" ]
null
null
Bundle discounts are used by retailers in many industries. Optimal bundle pricing requires learning the joint distribution of consumer valuations for the items in the bundle, that is, how much they are willing to pay for each of the items. We suppose that a retailer has sales transaction data, and the corresponding con...
[]
null
25
null
null
[ 0.02309391275048256, -0.019110247492790222, -0.008832788094878197, 0.05173706263303757, 0.06306102126836777, 0.06027720868587494, 0.004419206641614437, -0.017752349376678467, 0.006369506008923054, -0.02763235755264759, 0.005403113551437855, 0.020876916125416756, -0.05296628177165985, 0.019...
Towards Minimax Online Learning with Unknown Time Horizon
https://proceedings.mlr.press/v32/luo14.html
[ "Haipeng Luo", "Robert Schapire" ]
null
null
We consider online learning when the time horizon is unknown. We apply a minimax analysis, beginning with the fixed horizon case, and then moving on to two unknown-horizon settings, one that assumes the horizon is chosen randomly according to some distribution, and the other which allows the adversary full control over...
[]
null
26
1307.8187
title_snapshot
[ -0.03636404499411583, -0.014854400418698788, 0.01272794883698225, 0.0448453351855278, 0.04250321909785271, 0.03656081110239029, 0.010081568732857704, 0.016611561179161072, -0.021450933068990707, -0.03058677352964878, -0.03753140568733215, 0.008204414509236813, -0.05445725843310356, -0.0204...
Factorized Point Process Intensities: A Spatial Analysis of Professional Basketball
https://proceedings.mlr.press/v32/miller14.html
[ "Andrew Miller", "Luke Bornn", "Ryan Adams", "Kirk Goldsberry" ]
null
null
We develop a machine learning approach to represent and analyze the underlying spatial structure that governs shot selection among professional basketball players in the NBA. Typically, NBA players are discussed and compared in an heuristic, imprecise manner that relies on unmeasured intuitions about player behavior. ...
[]
null
27
1401.0942
title_snapshot
[ 0.012715737335383892, 0.0030922445002943277, -0.018822859972715378, -0.010744272731244564, 0.06265778094530106, 0.019166983664035797, 0.04289492219686508, -0.01343956682831049, -0.011748310178518295, -0.07695475965738297, 0.00853573065251112, -0.03618514910340309, -0.08849412947893143, 0.0...
Margins, Kernels and Non-linear Smoothed Perceptrons
https://proceedings.mlr.press/v32/ramdas14.html
[ "Aaditya Ramdas", "Javier Peña" ]
null
null
We focus on the problem of finding a non-linear classification function that lies in a Reproducing Kernel Hilbert Space (RKHS) both from the primal point of view (finding a perfect separator when one exists) and the dual point of view (giving a certificate of non-existence), with special focus on generalizations of two...
[]
null
28
1505.04123
title_snapshot
[ -0.02690288983285427, 0.007223988883197308, 0.03337470069527626, 0.041791003197431564, 0.014478659257292747, 0.045401301234960556, 0.01300150528550148, -0.009274797514081001, -0.039730530232191086, -0.04259100928902626, -0.023292895406484604, 0.021948043256998062, -0.05505891144275665, -0....
Robust RegBayes: Selectively Incorporating First-Order Logic Domain Knowledge into Bayesian Models
https://proceedings.mlr.press/v32/mei14.html
[ "Shike Mei", "Jun Zhu", "Jerry Zhu" ]
null
null
Much research in Bayesian modeling has been done to elicit a prior distribution that incorporates domain knowledge. We present a novel and more direct approach by imposing First-Order Logic (FOL) rules on the posterior distribution. Our approach unifies FOL and Bayesian modeling under the regularized Bayesian framework...
[]
null
29
null
null
[ 0.014997242018580437, 0.0073370253667235374, -0.018829721957445145, 0.028922371566295624, 0.07069534063339233, -0.0050481464713811874, 0.02713722735643387, -0.008925122208893299, -0.039818938821554184, -0.010145231150090694, -0.01942080445587635, 0.047052495181560516, -0.07487363368272781, ...
Learning Theory and Algorithms for revenue optimization in second price auctions with reserve
https://proceedings.mlr.press/v32/mohri14.html
[ "Mehryar Mohri", "Andres Munoz Medina" ]
null
null
Second-price auctions with reserve play a critical role for modern search engine and popular online sites since the revenue of these companies often directly depends on the outcome of such auctions. The choice of the reserve price is the main mechanism through which the auction revenue can be influenced in the...
[]
null
30
1310.5665
title_snapshot
[ -0.04307568073272705, -0.018530404195189476, 0.006656705401837826, 0.018851827830076218, 0.025986284017562866, 0.017959319055080414, 0.00013243843568488955, 0.011943521909415722, -0.020298514515161514, -0.024673718959093094, -0.005984645336866379, 0.01888195611536503, -0.042131561785936356, ...
Low-density Parity Constraints for Hashing-Based Discrete Integration
https://proceedings.mlr.press/v32/ermon14.html
[ "Stefano Ermon", "Carla Gomes", "Ashish Sabharwal", "Bart Selman" ]
null
null
In recent years, a number of probabilistic inference and counting techniques have been proposed that exploit pairwise independent hash functions to infer properties of succinctly defined high-dimensional sets. While providing desirable statistical guarantees, typical constructions of such hash functions are themselves ...
[]
null
31
null
null
[ -0.01982479728758335, 0.0031156891491264105, -0.03086584433913231, 0.06960073858499527, 0.03749668225646019, 0.011710064485669136, 0.04089774563908577, 0.0015003543812781572, -0.018129862844944, -0.023116549476981163, 0.0017252765828743577, -0.0516078919172287, -0.0580858439207077, -0.0187...
Prediction with Limited Advice and Multiarmed Bandits with Paid Observations
https://proceedings.mlr.press/v32/seldin14.html
[ "Yevgeny Seldin", "Peter Bartlett", "Koby Crammer", "Yasin Abbasi-Yadkori" ]
null
null
We study two problems of online learning under restricted information access. In the first problem, \emphprediction with limited advice, we consider a game of prediction with expert advice, where on each round of the game we query the advice of a subset of M out of N experts. We present an algorithm that achieves O(\sq...
[]
null
32
null
null
[ -0.017984231933951378, -0.003806223627179861, -0.0006084626074880362, 0.035422585904598236, 0.04445015266537666, 0.019877292215824127, 0.027510561048984528, 0.000020487104848143645, -0.023891065269708633, -0.03502571955323219, -0.017765263095498085, 0.028499390929937363, -0.06500499695539474...
Bayesian Nonparametric Multilevel Clustering with Group-Level Contexts
https://proceedings.mlr.press/v32/nguyenb14.html
[ "Tien Vu Nguyen", "Dinh Phung", "Xuanlong Nguyen", "Swetha Venkatesh", "Hung Bui" ]
null
null
We present a Bayesian nonparametric framework for multilevel clustering which utilizes group-level context information to simultaneously discover low-dimensional structures of the group contents and partitions groups into clusters. Using the Dirichlet process as the building block, our model constructs a product base-m...
[]
null
33
1401.1974
title_snapshot
[ 0.0012814328074455261, 0.01804962381720543, 0.014659295789897442, 0.04252983629703522, 0.0075456248596310616, 0.028151588514447212, 0.022321801632642746, 0.002685325685888529, -0.0029424019157886505, -0.046826090663671494, -0.020347869023680687, -0.003877977840602398, -0.033698927611112595, ...
Large-Margin Metric Learning for Constrained Partitioning Problems
https://proceedings.mlr.press/v32/lajugie14.html
[ "Rémi Lajugie", "Francis Bach", "Sylvain Arlot" ]
null
null
We consider unsupervised partitioning problems based explicitly or implicitly on the minimization of Euclidean distortions, such as clustering, image or video segmentation, and other change-point detection problems. We emphasize on cases with specific structure, which include many practical situations ranging from mean...
[]
null
34
null
null
[ -0.010154674760997295, -0.02926384098827839, -0.032355908304452896, 0.02359158545732498, 0.04097364470362663, 0.021768903359770775, 0.011204400099813938, 0.00260316114872694, -0.0281693022698164, -0.029405780136585236, -0.011433701030910015, -0.002341807819902897, -0.05815577507019043, 0.0...
Wasserstein Propagation for Semi-Supervised Learning
https://proceedings.mlr.press/v32/solomon14.html
[ "Justin Solomon", "Raif Rustamov", "Leonidas Guibas", "Adrian Butscher" ]
null
null
Probability distributions and histograms are natural representations for product ratings, traffic measurements, and other data considered in many machine learning applications. Thus, this paper introduces a technique for graph-based semi-supervised learning of histograms, derived from the theory of optimal transportat...
[]
null
35
null
null
[ -0.009642348624765873, -0.035606108605861664, 0.0211931262165308, 0.04135291650891304, 0.034869518131017685, 0.026863770559430122, 0.002609502524137497, 0.01133895292878151, -0.0027031556237488985, -0.06279028207063675, -0.014183753170073032, -0.006852780934423208, -0.06668100506067276, 0....
Max-Margin Infinite Hidden Markov Models
https://proceedings.mlr.press/v32/zhangb14.html
[ "Aonan Zhang", "Jun Zhu", "Bo Zhang" ]
null
null
Infinite hidden Markov models (iHMMs) are nonparametric Bayesian extensions of hidden Markov models (HMMs) with an infinite number of states. Though flexible in describing sequential data, the generative formulation of iHMMs could limit their discriminative ability in sequential prediction tasks. Our paper introduces m...
[]
null
36
null
null
[ -0.007349926047027111, 0.020196612924337387, -0.00010172996553592384, 0.026969611644744873, 0.04637942835688591, 0.009151014499366283, 0.05686495825648308, 0.010732618160545826, -0.034696102142333984, -0.04398762062191963, 0.012045183219015598, 0.012855862267315388, -0.06894360482692719, 0...
Efficient Approximation of Cross-Validation for Kernel Methods using Bouligand Influence Function
https://proceedings.mlr.press/v32/liua14.html
[ "Yong Liu", "Shali Jiang", "Shizhong Liao" ]
null
null
Model selection is one of the key issues both in recent research and application of kernel methods. Cross-validation is a commonly employed and widely accepted model selection criterion. However, it requires multiple times of training the algorithm under consideration, which is computationally intensive. In this paper,...
[]
null
37
null
null
[ -0.04900381714105606, -0.028812840580940247, 0.024222398176789284, -0.00535850552842021, 0.02433021180331707, 0.02077607251703739, 0.014422272332012653, -0.02292395383119583, -0.012842626310884953, -0.021563494578003883, 0.00884808786213398, 0.0385153591632843, -0.04454571381211281, 0.0164...
Generalized Exponential Concentration Inequality for Renyi Divergence Estimation
https://proceedings.mlr.press/v32/singh14.html
[ "Shashank Singh", "Barnabas Poczos" ]
null
null
Estimating divergences between probability distributions in a consistent way is of great importance in many machine learning tasks. Although this is a fundamental problem in nonparametric statistics, to the best of our knowledge there has been no finite sample exponential inequality convergence bound derived for any di...
[]
null
38
1603.08589
title_snapshot
[ -0.02200503833591938, -0.0035637877881526947, 0.0007747630588710308, 0.021762747317552567, 0.04196355119347572, 0.04576350376009941, 0.020021751523017883, -0.0035204198211431503, -0.029637744650244713, -0.04205930605530739, 0.007403201423585415, -0.0000345656335412059, -0.08225452154874802, ...
Boosting with Online Binary Learners for the Multiclass Bandit Problem
https://proceedings.mlr.press/v32/chenb14.html
[ "Shang-Tse Chen", "Hsuan-Tien Lin", "Chi-Jen Lu" ]
null
null
We consider the problem of online multiclass prediction in the bandit setting. Compared with the full-information setting, in which the learner can receive the true label as feedback after making each prediction, the bandit setting assumes that the learner can only know the correctness of the predicted label. Because t...
[]
null
39
null
null
[ -0.007989449426531792, -0.02338973805308342, -0.016118410974740982, 0.05728979408740997, 0.02668946608901024, 0.022143419831991196, 0.016241971403360367, 0.006386833731085062, -0.012617823667824268, -0.029897408559918404, -0.041077204048633575, 0.026342788711190224, -0.08056104928255081, -...
Optimal Budget Allocation: Theoretical Guarantee and Efficient Algorithm
https://proceedings.mlr.press/v32/soma14.html
[ "Tasuku Soma", "Naonori Kakimura", "Kazuhiro Inaba", "Ken-ichi Kawarabayashi" ]
null
null
We consider the budget allocation problem over bipartite influence model proposed by Alon et al. This problem can be viewed as the well-known influence maximization problem with budget constraints. We first show that this problem and its much more general form fall into a general setting; namely the monotone submo...
[]
null
40
null
null
[ -0.015668151900172234, -0.015401787124574184, -0.013960208743810654, 0.030868448317050934, 0.03488081321120262, 0.04029783979058266, 0.005400343798100948, -0.01164551917463541, -0.026602953672409058, -0.05297640338540077, -0.008188623003661633, -0.011932079680263996, -0.06882648915052414, ...
Computing Parametric Ranking Models via Rank-Breaking
https://proceedings.mlr.press/v32/soufiani14.html
[ "Hossein Azari Soufiani", "David Parkes", "Lirong Xia" ]
null
null
Rank breaking is a methodology introduced by Azari Soufiani et al. (2013a) for applying a Generalized Method of Moments (GMM) algorithm to the estimation of parametric ranking models. Breaking takes full rankings and breaks, or splits them up, into counts for pairs of alternatives that occur in particular positions (e....
[]
null
41
null
null
[ -0.040718063712120056, -0.030093783512711525, -0.02787702903151512, 0.018056746572256088, 0.010377095080912113, 0.0245375894010067, 0.01575281284749508, 0.013726454228162766, -0.02521171234548092, -0.05064614862203598, 0.0017998566618189216, 0.007423858158290386, -0.05261711776256561, -0.0...
Tracking Adversarial Targets
https://proceedings.mlr.press/v32/abbasi-yadkori14.html
[ "Yasin Abbasi-Yadkori", "Peter Bartlett", "Varun Kanade" ]
null
null
We study linear control problems with quadratic losses and adversarially chosen tracking targets. We present an efficient algorithm for this problem and show that, under standard conditions on the linear system, its regret with respect to an optimal linear policy grows as O(\log^2 T), where T is the number of rounds of...
[]
null
42
null
null
[ -0.033833134919404984, -0.01779252104461193, 0.0043114665895700455, 0.019640587270259857, 0.026943275704979897, 0.025809215381741524, 0.0072890110313892365, 0.008483190089464188, -0.022558989003300667, -0.05496864393353462, -0.039281751960515976, 0.007250515278428793, -0.06751322746276855, ...
Online Bayesian Passive-Aggressive Learning
https://proceedings.mlr.press/v32/shi14.html
[ "Tianlin Shi", "Jun Zhu" ]
null
null
Online Passive-Aggressive (PA) learning is an effective framework for performing max-margin online learning. But the deterministic formulation and estimated single large-margin model could limit its capability in discovering descriptive structures underlying complex data. This paper presents online Bayesian Passive-Agg...
[]
null
43
1312.3388
title_snapshot
[ 0.022340774536132812, -0.0003131239500362426, 0.014366383664309978, 0.02694425918161869, 0.008863070979714394, 0.02114754170179367, 0.03712755814194679, -0.019836775958538055, -0.004414274822920561, -0.008436782285571098, -0.005978666711598635, 0.013614613562822342, -0.06175551936030388, -...
Deterministic Policy Gradient Algorithms
https://proceedings.mlr.press/v32/silver14.html
[ "David Silver", "Guy Lever", "Nicolas Heess", "Thomas Degris", "Daan Wierstra", "Martin Riedmiller" ]
null
null
In this paper we consider deterministic policy gradient algorithms for reinforcement learning with continuous actions. The deterministic policy gradient has a particularly appealing form: it is the expected gradient of the action-value function. This simple form means that the deterministic policy gradient can be estim...
[]
null
44
null
null
[ -0.005941174924373627, -0.01990627869963646, 0.0010263246949762106, 0.035509590059518814, 0.02093731425702572, 0.040601275861263275, 0.034122008830308914, -0.0009155147708952427, -0.03575034439563751, -0.0256732739508152, -0.014384630136191845, 0.019181033596396446, -0.06740628182888031, -...
Modeling Correlated Arrival Events with Latent Semi-Markov Processes
https://proceedings.mlr.press/v32/lian14.html
[ "Wenzhao Lian", "Vinayak Rao", "Brian Eriksson", "Lawrence Carin" ]
null
null
The analysis and characterization of correlated point process data has wide applications, ranging from biomedical research to network analysis. In this work, we model such data as generated by a latent collection of continuous-time binary semi-Markov processes, corresponding to external events appearing and disappearin...
[]
null
45
null
null
[ 0.014434226788580418, -0.014518833719193935, -0.03320587798953056, 0.023548096418380737, 0.049287229776382446, 0.03480280190706253, 0.02237592078745365, 0.01817711628973484, -0.014868770726025105, -0.050080351531505585, 0.006270825397223234, -0.024290449917316437, -0.058375339955091476, 0....
Towards scaling up Markov chain Monte Carlo: an adaptive subsampling approach
https://proceedings.mlr.press/v32/bardenet14.html
[ "Rémi Bardenet", "Arnaud Doucet", "Chris Holmes" ]
null
null
Markov chain Monte Carlo (MCMC) methods are often deemed far too computationally intensive to be of any practical use for large datasets. This paper describes a methodology that aims to scale up the Metropolis-Hastings (MH) algorithm in this context. We propose an approximate implementation of the accept/reject step of...
[]
null
46
null
null
[ -0.010240266099572182, -0.018983779475092888, -0.024878934025764465, 0.024735227227211, 0.07315046340227127, 0.039354871958494186, 0.04076926037669182, -0.02356881834566593, -0.01523527316749096, -0.04976561665534973, 0.019506316632032394, -0.00026164815062657, -0.06711379438638687, 0.0032...
Diagnosis determination: decision trees optimizing simultaneously worst and expected testing cost
https://proceedings.mlr.press/v32/cicalese14.html
[ "Ferdinando Cicalese", "Eduardo Laber", "Aline Medeiros Saettler" ]
null
null
In several applications of automatic diagnosis and active learning a central problem is the evaluation of a discrete function by adaptively querying the values of its variables until the values read uniquely determine the value of the function. In general reading the value of a variable is done at the expense of som...
[]
null
47
1309.2796
title_judge
[ -0.03803358972072601, 0.020385785028338432, -0.027460703626275063, 0.028430838137865067, 0.040024254471063614, 0.04464627802371979, 0.012970647774636745, -0.024729784578084946, 0.011342114768922329, -0.022611893713474274, -0.004241174552589655, 0.032436009496450424, -0.04637930542230606, -...
Condensed Filter Tree for Cost-Sensitive Multi-Label Classification
https://proceedings.mlr.press/v32/lia14.html
[ "Chun-Liang Li", "Hsuan-Tien Lin" ]
null
null
Different real-world applications of multi-label classification often demand different evaluation criteria. We formalize this demand with a general setup, cost-sensitive multi-label classification (CSMLC), which takes the evaluation criteria into account during learning. Nevertheless, most existing algorithms can only...
[]
null
48
null
null
[ -0.00856050755828619, -0.02588564343750477, 0.010431134141981602, 0.036232903599739075, 0.05467919632792473, 0.009619529359042645, -0.010154773481190205, -0.0133656682446599, -0.0037331683561205864, -0.018040448427200317, -0.017255015671253204, -0.0023920743260532618, -0.0766448900103569, ...
On Measure Concentration of Random Maximum A-Posteriori Perturbations
https://proceedings.mlr.press/v32/orabona14.html
[ "Francesco Orabona", "Tamir Hazan", "Anand Sarwate", "Tommi Jaakkola" ]
null
null
The maximum a-posteriori (MAP) perturbation framework has emerged as a useful approach for inference and learning in high dimensional complex models. By maximizing a randomly perturbed potential function, MAP perturbations generate unbiased samples from the Gibbs distribution. Unfortunately, the computational cost of...
[]
null
49
1310.4227
title_snapshot
[ -0.036034662276506424, -0.009777777828276157, 0.014374131336808205, 0.015061457641422749, 0.01258792169392109, 0.030347531661391258, 0.030725156888365746, -0.01461756881326437, -0.0198893491178751, -0.04738673195242882, 0.008207893930375576, -0.020710807293653488, -0.05261602625250816, -0....
Bias in Natural Actor-Critic Algorithms
https://proceedings.mlr.press/v32/thomas14.html
[ "Philip Thomas" ]
null
null
We show that several popular discounted reward natural actor-critics, including the popular NAC-LSTD and eNAC algorithms, do not generate unbiased estimates of the natural policy gradient as claimed. We derive the first unbiased discounted reward natural actor-critics using batch and iterative approaches to gradient es...
[]
null
50
null
null
[ -0.021596068516373634, -0.04148878529667854, -0.015118518844246864, 0.04834151640534401, 0.016758430749177933, 0.04052390903234482, 0.027452120557427406, 0.014469764195382595, -0.031546451151371, -0.030824722722172737, -0.0021577172446995974, 0.041163280606269836, -0.07746435701847076, -0....
Dimension-free Concentration Bounds on Hankel Matrices for Spectral Learning
https://proceedings.mlr.press/v32/denis14.html
[ "François Denis", "Mattias Gybels", "Amaury Habrard" ]
null
null
Learning probabilistic models over strings is an important issue for many applications. Spectral methods propose elegant solutions to the problem of inferring weighted automata from finite samples of variable-length strings drawn from an unknown target distribution. These methods rely on a singular value decomposition ...
[]
null
51
1312.6282
title_snapshot
[ -0.03775901719927788, 0.021086277440190315, -0.006339983083307743, 0.023054566234350204, 0.032789140939712524, 0.026075521484017372, 0.04711615666747093, 0.01638774760067463, -0.010703708045184612, -0.03406519815325737, -0.003096336731687188, -0.000212477330933325, -0.0822625607252121, 0.0...
On Modelling Non-linear Topical Dependencies
https://proceedings.mlr.press/v32/lib14.html
[ "Zhixing Li", "Siqiang Wen", "Juanzi Li", "Peng Zhang", "Jie Tang" ]
null
null
Probabilistic topic models such as Latent Dirichlet Allocation (LDA) discover latent topics from large corpora by exploiting words’ co-occurring relation. By observing the topical similarity between words, we find that some other relations, such as semantic or syntax relation between words, lead to strong dependence be...
[]
null
52
null
null
[ 0.017508987337350845, -0.02041403390467167, 0.014545894227921963, 0.04703157767653465, 0.01996930129826069, 0.022075941786170006, -0.0029843710362911224, 0.016403893008828163, -0.0008618527790531516, -0.003649937454611063, -0.034322112798690796, 0.035971954464912415, -0.06794887781143188, ...
A Deep and Tractable Density Estimator
https://proceedings.mlr.press/v32/uria14.html
[ "Benigno Uria", "Iain Murray", "Hugo Larochelle" ]
null
null
The Neural Autoregressive Distribution Estimator (NADE) and its real-valued version RNADE are competitive density models of multidimensional data across a variety of domains. These models use a fixed, arbitrary ordering of the data dimensions. One can easily condition on variables at the beginning of the ordering, and...
[]
null
53
1310.1757
title_snapshot
[ -0.024550801143050194, -0.0007950243889354169, -0.03756704181432724, 0.02464551106095314, 0.014496264979243279, 0.06502974778413773, 0.016332343220710754, -0.007922417484223843, -0.015497690066695213, -0.07466740906238556, 0.014178243465721607, 0.012628452852368355, -0.04626087471842766, 0...
(Near) Dimension Independent Risk Bounds for Differentially Private Learning
https://proceedings.mlr.press/v32/jain14.html
[ "Prateek Jain", "Abhradeep Guha Thakurta" ]
null
null
In this paper, we study the problem of differentially private risk minimization where the goal is to provide differentially private algorithms that have small excess risk. In particular we address the following open problem: \emphIs it possible to design computationally efficient differentially private risk minimizers ...
[]
null
54
null
null
[ -0.024519450962543488, 0.00955013558268547, 0.007414204068481922, 0.043190501630306244, 0.041141360998153687, 0.02995879575610161, 0.05097806453704834, -0.03597290441393852, -0.013955678790807724, -0.03154049813747406, -0.0007886815001256764, -0.005446037743240595, -0.06554342061281204, 0....
Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels
https://proceedings.mlr.press/v32/yangb14.html
[ "Jiyan Yang", "Vikas Sindhwani", "Haim Avron", "Michael Mahoney" ]
null
null
We consider the problem of improving the efficiency of randomized Fourier feature maps to accelerate training and testing speed of kernel methods on large datasets. These approximate feature maps arise as Monte Carlo approximations to integral representations of shift-invariant kernel functions (e.g., Gaussian kernel)....
[]
null
55
1412.8293
title_snapshot
[ -0.020820284262299538, -0.009155270643532276, 0.004603453446179628, 0.0550764761865139, 0.031596701592206955, 0.055382367223501205, 0.005008589942008257, 0.0010189437307417393, -0.030245501548051834, -0.0584295280277729, -0.033845312893390656, 0.006568163633346558, -0.06171135976910591, -0...
Discriminative Features via Generalized Eigenvectors
https://proceedings.mlr.press/v32/karampatziakis14.html
[ "Nikos Karampatziakis", "Paul Mineiro" ]
null
null
Representing examples in a way that is compatible with the underlying classifier can greatly enhance the performance of a learning system. In this paper we investigate scalable techniques for inducing discriminative features by taking advantage of simple second order structure in the data. We focus on multiclass classi...
[]
null
56
1310.1934
title_snapshot
[ -0.008661113679409027, -0.02111351490020752, 0.022780481725931168, 0.02449173294007778, 0.029699699953198433, 0.04167035222053528, 0.014647106640040874, -0.007573072798550129, -0.010646813549101353, -0.0447315089404583, -0.01714240573346615, 0.0228829775005579, -0.09300579130649567, 0.0144...
Forward-Backward Greedy Algorithms for General Convex Smooth Functions over A Cardinality Constraint
https://proceedings.mlr.press/v32/liub14.html
[ "Ji Liu", "Jieping Ye", "Ryohei Fujimaki" ]
null
null
We consider forward-backward greedy algorithms for solving sparse feature selection problems with general convex smooth functions. A state-of-the-art greedy method, the Forward-Backward greedy algorithm (FoBa-obj) requires to solve a large number of optimization problems, thus it is not scalable for large-size problems...
[]
null
57
1401.0086
title_snapshot
[ -0.03729545325040817, 0.0016605723649263382, 0.03438709303736687, 0.006207913625985384, 0.05101495608687401, 0.039251796901226044, 0.028365960344672203, -0.0010254731168970466, -0.0330672562122345, -0.041847165673971176, -0.030528979375958443, 0.015154295600950718, -0.07767987251281738, -0...
Online Learning in Markov Decision Processes with Changing Cost Sequences
https://proceedings.mlr.press/v32/dick14.html
[ "Travis Dick", "Andras Gyorgy", "Csaba Szepesvari" ]
null
null
In this paper we consider online learning in finite Markov decision processes (MDPs) with changing cost sequences under full and bandit-information. We propose to view this problem as an instance of online linear optimization. We propose two methods for this problem: MD^2 (mirror descent with approximate projections)...
[]
null
58
null
null
[ -0.052409522235393524, -0.006892716977745295, -0.0037331220228224993, 0.04081866517663002, 0.062304917722940445, 0.027421582490205765, 0.003510292386636138, 0.024032825604081154, -0.006360698025673628, -0.03371309116482735, -0.022513655945658684, 0.005080398637801409, -0.06985035538673401, ...
Unimodal Bandits: Regret Lower Bounds and Optimal Algorithms
https://proceedings.mlr.press/v32/combes14.html
[ "Richard Combes", "Alexandre Proutiere" ]
null
null
We consider stochastic multi-armed bandits where the expected reward is a unimodal function over partially ordered arms. This important class of problems has been recently investigated in (Cope 2009, Yu 2011). The set of arms is either discrete, in which case arms correspond to the vertices of a finite graph whose stru...
[]
null
59
1405.5096
title_snapshot
[ -0.031519073992967606, -0.024693677201867104, -0.013727515004575253, 0.038293395191431046, 0.0364081971347332, 0.028876645490527153, 0.027261171489953995, 0.03460768237709999, -0.01097237877547741, -0.05083618313074112, 0.007397848181426525, -0.0006700858939439058, -0.07171288877725601, -0...
Maximum Mean Discrepancy for Class Ratio Estimation: Convergence Bounds and Kernel Selection
https://proceedings.mlr.press/v32/iyer14.html
[ "Arun Iyer", "Saketha Nath", "Sunita Sarawagi" ]
null
null
In recent times, many real world applications have emerged that require estimates of class ratios in an unlabeled instance collection as opposed to labels of individual instances in the collection. In this paper we investigate the use of maximum mean discrepancy (MMD) in a reproducing kernel Hilbert space (RKHS) for e...
[]
null
60
null
null
[ -0.029798215255141258, 0.009969818405807018, -0.008890856057405472, 0.05032654479146004, 0.022161658853292465, 0.04117744415998459, 0.03663862124085426, -0.025361791253089905, -0.032629549503326416, -0.031509991735219955, -0.034095510840415955, 0.02465854398906231, -0.06695454567670822, 0....
Asymptotically consistent estimation of the number of change points in highly dependent time series
https://proceedings.mlr.press/v32/khaleghi14.html
[ "Azadeh Khaleghi", "Daniil Ryabko" ]
null
null
The problem of change point estimation is considered in a general framework where the data are generated by arbitrary unknown stationary ergodic process distributions. This means that the data may have long-range dependencies of an arbitrary form. In this context the consistent estimation of the number of change poin...
[]
null
61
1302.3407
title_judge
[ -0.02088889665901661, -0.03559955582022667, -0.02545405924320221, -0.012843307107686996, 0.06155138090252876, 0.06566130369901657, 0.026642512530088425, 0.02032654546201229, -0.02914663590490818, -0.052481818944215775, -0.0049363612197339535, 0.008333653211593628, -0.050618138164281845, 0....
Coordinate-descent for learning orthogonal matrices through Givens rotations
https://proceedings.mlr.press/v32/shalit14.html
[ "Uri Shalit", "Gal Chechik" ]
null
null
Optimizing over the set of orthogonal matrices is a central component in problems like sparse-PCA or tensor decomposition. Unfortunately, such optimization is hard since simple operations on orthogonal matrices easily break orthogonality, and correcting orthogonality usually costs a large amount of computation. Here w...
[]
null
62
1312.0624
title_judge
[ -0.027871035039424896, -0.021920684725046158, 0.0010114273754879832, 0.022992201149463654, 0.018269937485456467, 0.028510242700576782, 0.022029032930731773, 0.016648543998599052, -0.025345342233777046, -0.037192072719335556, -0.011391621083021164, 0.0025629426818341017, -0.07716134190559387,...
Densifying One Permutation Hashing via Rotation for Fast Near Neighbor Search
https://proceedings.mlr.press/v32/shrivastava14.html
[ "Anshumali Shrivastava", "Ping Li" ]
null
null
The query complexity of \em locality sensitive hashing (LSH) based similarity search is dominated by the number of hash evaluations, and this number grows with the data size \citeProc:Indyk_STOC98. In industrial applications such as search where the data are often high-dimensional and binary (e.g., text n-grams), \em ...
[]
null
63
null
null
[ -0.03455758094787598, 0.0003454816760495305, -0.019352136179804802, 0.030919387936592102, 0.037647753953933716, 0.032368894666433334, 0.030577948316931725, 0.002249296987429261, -0.018031474202871323, -0.053755421191453934, 0.006138569675385952, -0.06055619195103645, -0.05840069428086281, ...
A Divide-and-Conquer Solver for Kernel Support Vector Machines
https://proceedings.mlr.press/v32/hsieha14.html
[ "Cho-Jui Hsieh", "Si Si", "Inderjit Dhillon" ]
null
null
The kernel support vector machine (SVM) is one of the most widely used classification methods; however, the amount of computation required becomes the bottleneck when facing millions of samples. In this paper, we propose and analyze a novel divide-and-conquer solver for kernel SVMs (DC-SVM). In the division step, we pa...
[]
null
64
1311.0914
title_snapshot
[ -0.03614203259348869, -0.02358611300587654, 0.007337003946304321, 0.04168761894106865, 0.029842428863048553, 0.047257695347070694, 0.006904235575348139, -0.044629842042922974, -0.029281795024871826, -0.01719922199845314, -0.00888893660157919, 0.00844250712543726, -0.06300004571676254, 0.03...
Nuclear Norm Minimization via Active Subspace Selection
https://proceedings.mlr.press/v32/hsiehb14.html
[ "Cho-Jui Hsieh", "Peder Olsen" ]
null
null
We describe a novel approach to optimizing matrix problems involving nuclear norm regularization and apply it to the matrix completion problem. We combine methods from non-smooth and smooth optimization. At each step we use the proximal gradient to select an active subspace. We then find a smooth, convex relaxation of ...
[]
null
65
null
null
[ -0.023203182965517044, -0.02238202653825283, 0.012248926796019077, 0.021752120926976204, 0.03035997971892357, -0.0033263908699154854, 0.019665274769067764, -0.025511624291539192, -0.03101508505642414, -0.0494307205080986, -0.026789292693138123, 0.0014569087652489543, -0.04135218262672424, ...
Provable Bounds for Learning Some Deep Representations
https://proceedings.mlr.press/v32/arora14.html
[ "Sanjeev Arora", "Aditya Bhaskara", "Rong Ge", "Tengyu Ma" ]
null
null
We give algorithms with provable guarantees that learn a class of deep nets in the generative model view popularized by Hinton and others. Our generative model is an n node multilayer neural net that has degree at most n^γ for some γ< 1 and each edge has a random edge weight in [-1,1]. Our algorithm learns almost all...
[]
null
66
1310.6343
title_snapshot
[ -0.014745160937309265, -0.01698552444577217, -0.001682730158790946, 0.059908654540777206, 0.03230917453765869, 0.004350789822638035, 0.026739919558167458, -0.00467961048707366, -0.02012983150780201, -0.02196599170565605, -0.007516857236623764, -0.035101454704999924, -0.06777084618806839, 0...
Large-scale Multi-label Learning with Missing Labels
https://proceedings.mlr.press/v32/yu14.html
[ "Hsiang-Fu Yu", "Prateek Jain", "Purushottam Kar", "Inderjit Dhillon" ]
null
null
The multi-label classification problem has generated significant interest in recent years. However, existing approaches do not adequately address two key challenges: (a) scaling up to problems with a large number (say millions) of labels, and (b) handling data with missing labels. In this paper, we directly address bot...
[]
null
67
1307.5101
title_snapshot
[ -0.027749525383114815, -0.03852248191833496, -0.005972781218588352, 0.041410185396671295, 0.023534033447504044, 0.03041139803826809, 0.0011187740601599216, -0.02884490415453911, -0.04183783382177353, -0.029397524893283844, -0.010966366156935692, 0.022990690544247627, -0.07167033851146698, ...
Learning Graphs with a Few Hubs
https://proceedings.mlr.press/v32/tandon14.html
[ "Rashish Tandon", "Pradeep Ravikumar" ]
null
null
We consider the problem of recovering the graph structure of a “hub-networked” Ising model given iid samples, under high-dimensional settings, where number of nodes p could be potentially larger than the number of samples n. By a “hub-networked” graph, we mean a graph with a few “hub nodes” with very large degrees. Sta...
[]
null
68
null
null
[ -0.02604767121374607, -0.027823159471154213, -0.0019337700214236975, 0.05071549117565155, 0.041075609624385834, 0.004681775812059641, 0.027085820212960243, 0.03441890329122543, -0.01836625300347805, -0.041651155799627304, 0.019600283354520798, -0.034938592463731766, -0.07683955132961273, 0...
Agnostic Bayesian Learning of Ensembles
https://proceedings.mlr.press/v32/lacoste14.html
[ "Alexandre Lacoste", "Mario Marchand", "François Laviolette", "Hugo Larochelle" ]
null
null
We propose a method for producing ensembles of predictors based on holdout estimations of their generalization performances. This approach uses a prior directly on the performance of predictors taken from a finite set of candidates and attempts to infer which one is best. Using Bayesian inference, we can thus obtain a ...
[]
null
69
null
null
[ -0.01221309881657362, -0.0015164508949965239, -0.013389002531766891, 0.029955320060253143, 0.037410251796245575, 0.025606347247958183, 0.026056641712784767, -0.04977884143590927, -0.03764214739203453, -0.029919365420937538, -0.021235371008515358, 0.020937247201800346, -0.08252906799316406, ...
Towards an optimal stochastic alternating direction method of multipliers
https://proceedings.mlr.press/v32/azadi14.html
[ "Samaneh Azadi", "Suvrit Sra" ]
null
null
We study regularized stochastic convex optimization subject to linear equality constraints. This class of problems was recently also studied by Ouyang et al. (2013) and Suzuki (2013); both introduced similar stochastic alternating direction method of multipliers (SADMM) algorithms. However, the analysis of both papers ...
[]
null
70
null
null
[ -0.028335511684417725, 0.01016435120254755, -0.0017628323985263705, 0.007516026962548494, 0.02494499832391739, 0.07310912758111954, 0.035403717309236526, 0.009477770887315273, -0.05855615437030792, -0.03918773680925369, -0.0020402870140969753, -0.02297051064670086, -0.04364917427301407, -0...
Spherical Hamiltonian Monte Carlo for Constrained Target Distributions
https://proceedings.mlr.press/v32/lan14.html
[ "Shiwei Lan", "Bo Zhou", "Babak Shahbaba" ]
null
null
Statistical models with constrained probability distributions are abundant in machine learning. Some examples include regression models with norm constraints (e.g., Lasso), probit models, many copula models, and Latent Dirichlet Allocation (LDA) models. Bayesian inference involving probability distributions confined to...
[]
null
71
1309.4289
title_snapshot
[ -0.026691019535064697, 0.028759511187672615, -0.012981876730918884, 0.031162573024630547, 0.03035060502588749, 0.022997630760073662, 0.015190675854682922, -0.019412504509091377, -0.03454149141907692, -0.04539031162858009, -0.01048686821013689, 0.012632458470761776, -0.07216763496398926, 0....
Efficient Continuous-Time Markov Chain Estimation
https://proceedings.mlr.press/v32/hajiaghayi14.html
[ "Monir Hajiaghayi", "Bonnie Kirkpatrick", "Liangliang Wang", "Alexandre Bouchard-Côté" ]
null
null
Many problems of practical interest rely on Continuous-time Markov chains (CTMCs) defined over combinatorial state spaces, rendering the computation of transition probabilities, and hence probabilistic inference, difficult or impossible with existing methods. For problems with countably infinite states, where classica...
[]
null
72
1309.3250
title_snapshot
[ -0.0007825389620848, 0.008129927329719067, -0.04762507230043411, 0.04059593379497528, 0.05157579481601715, 0.016940494999289513, 0.04899303615093231, 0.03819704055786133, -0.0027938561979681253, -0.05252574756741524, 0.04418305680155754, -0.0013964641839265823, -0.07899142056703568, -0.009...
DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
https://proceedings.mlr.press/v32/donahue14.html
[ "Jeff Donahue", "Yangqing Jia", "Oriol Vinyals", "Judy Hoffman", "Ning Zhang", "Eric Tzeng", "Trevor Darrell" ]
null
null
We evaluate whether features extracted from the activation of a deep convolutional network trained in a fully supervised fashion on a large, fixed set of object recognition tasks can be re-purposed to novel generic tasks. Our generic tasks may differ significantly from the originally trained tasks and there may be ins...
[]
null
73
1310.1531
title_snapshot
[ -0.000329331960529089, -0.024379340931773186, -0.02552851475775242, 0.02905459702014923, 0.03827957063913345, 0.020199505612254143, 0.03500468656420708, 0.04069565236568451, -0.017052503302693367, -0.033178944140672684, -0.03798038512468338, 0.005044958088546991, -0.06511884182691574, 0.02...
Making the Most of Bag of Words: Sentence Regularization with Alternating Direction Method of Multipliers
https://proceedings.mlr.press/v32/yogatama14.html
[ "Dani Yogatama", "Noah Smith" ]
null
null
In many high-dimensional learning problems, only some parts of an observation are important to the prediction task; for example, the cues to correctly categorizing a document may lie in a handful of its sentences. We introduce a learning algorithm that exploits this intuition by encoding it in a regularizer. Specifica...
[]
null
74
null
null
[ -0.03433156758546829, -0.03128562495112419, -0.006462001241743565, 0.017284059897065163, 0.030162150040268898, 0.021414460614323616, 0.03737867996096611, 0.0006932239630259573, -0.038841817528009415, -0.020792236551642418, -0.0076277051120996475, 0.009188800118863583, -0.06965018808841705, ...
Narrowing the Gap: Random Forests In Theory and In Practice
https://proceedings.mlr.press/v32/denil14.html
[ "Misha Denil", "David Matheson", "Nando De Freitas" ]
null
null
Despite widespread interest and practical use, the theoretical properties of random forests are still not well understood. In this paper we contribute to this understanding in two ways. We present a new theoreti- cally tractable variant of random regression forests and prove that our algorithm is con- sistent. We also ...
[]
null
75
1310.1415
title_snapshot
[ 0.022122398018836975, -0.03073653019964695, 0.01552466582506895, 0.04510509967803955, 0.036417145282030106, 0.044661007821559906, 0.03725825995206833, -0.0030093539971858263, -0.044955749064683914, -0.016739733517169952, -0.0009485141490586102, 0.0024456505198031664, -0.061665039509534836, ...
Coherent Matrix Completion
https://proceedings.mlr.press/v32/chenc14.html
[ "Yudong Chen", "Srinadh Bhojanapalli", "Sujay Sanghavi", "Rachel Ward" ]
null
null
Matrix completion concerns the recovery of a low-rank matrix from a subset of its revealed entries, and nuclear norm minimization has emerged as an effective surrogate for this combinatorial problem. Here, we show that nuclear norm minimization can recover an arbitrary n \times n matrix of rank r from O(nr log^2(n)) r...
[]
null
76
null
null
[ -0.018394803628325462, -0.012608975172042847, 0.017272617667913437, 0.03607016056776047, 0.03469362482428551, -0.006325359921902418, 0.017429538071155548, -0.022830264642834663, -0.03907403722405434, -0.04941108822822571, -0.0069449180737137794, -0.006740403827279806, -0.041538119316101074, ...
Admixture of Poisson MRFs: A Topic Model with Word Dependencies
https://proceedings.mlr.press/v32/inouye14.html
[ "David Inouye", "Pradeep Ravikumar", "Inderjit Dhillon" ]
null
null
This paper introduces a new topic model based on an admixture of Poisson Markov Random Fields (APM), which can model dependencies between words as opposed to previous independent topic models such as PLSA (Hofmann, 1999), LDA (Blei et al., 2003) or SAM (Reisinger et al., 2010). We propose a class of admixture models th...
[]
null
77
null
null
[ -0.001621020259335637, -0.018760401755571365, -0.03399467095732689, 0.041812870651483536, 0.036548200994729996, 0.05894550308585167, 0.02540801279246807, 0.02235819771885872, -0.029049504548311234, -0.023548917844891548, -0.013382983393967152, -0.00024252611910924315, -0.08450279384851456, ...
True Online TD(lambda)
https://proceedings.mlr.press/v32/seijen14.html
[ "Harm Seijen", "Rich Sutton" ]
null
null
TD(lambda) is a core algorithm of modern reinforcement learning. Its appeal comes from its equivalence to a clear and conceptually simple forward view, and the fact that it can be implemented online in an inexpensive manner. However, the equivalence between TD(lambda) and the forward view is exact only for the off-line...
[]
null
78
null
null
[ -0.03645046427845955, -0.02582700364291668, 0.0032030593138188124, 0.006803113501518965, 0.03214814141392708, 0.014600967057049274, 0.004729073960334063, 0.005313195753842592, -0.02546439692378044, -0.013469908386468887, -0.007480005268007517, 0.01332953479140997, -0.0667954608798027, -0.0...
Memory Efficient Kernel Approximation
https://proceedings.mlr.press/v32/si14.html
[ "Si Si", "Cho-Jui Hsieh", "Inderjit Dhillon" ]
null
null
The scalability of kernel machines is a big challenge when facing millions of samples due to storage and computation issues for large kernel matrices, that are usually dense. Recently, many papers have suggested tackling this problem by using a low rank approximation of the kernel matrix. In this paper, we first make t...
[]
null
79
null
null
[ -0.05726278945803642, -0.035631366074085236, 0.020057925954461098, 0.03443193435668945, 0.03432806581258774, 0.046030256897211075, 0.014109940268099308, -0.01803317293524742, -0.04383822903037071, -0.014679440297186375, 0.014081567525863647, 0.004629644099622965, -0.053461700677871704, 0.0...
Learning Sum-Product Networks with Direct and Indirect Variable Interactions
https://proceedings.mlr.press/v32/rooshenas14.html
[ "Amirmohammad Rooshenas", "Daniel Lowd" ]
null
null
Sum-product networks (SPNs) are a deep probabilistic representation that allows for efficient, exact inference. SPNs generalize many other tractable models, including thin junction trees, latent tree models, and many types of mixtures. Previous work on learning SPN structure has mainly focused on using top-down or bo...
[]
null
80
null
null
[ 0.011331242509186268, 0.019366059452295303, 0.015051905065774918, 0.026687659323215485, 0.041215717792510986, 0.020849989727139473, 0.033045269548892975, -0.013634526170790195, -0.016799207776784897, -0.010084440931677818, 0.01145991776138544, 0.01714286021888256, -0.07318460941314697, 0.0...
Hamiltonian Monte Carlo Without Detailed Balance
https://proceedings.mlr.press/v32/sohl-dickstein14.html
[ "Jascha Sohl-Dickstein", "Mayur Mudigonda", "Michael DeWeese" ]
null
null
We present a method for performing Hamiltonian Monte Carlo that largely eliminates sample rejection. In situations that would normally lead to rejection, instead a longer trajectory is computed until a new state is reached that can be accepted. This is achieved using Markov chain transitions that satisfy the fixed po...
[]
null
81
1409.5191
title_snapshot
[ -0.012982441112399101, 0.01178760640323162, -0.03270384669303894, 0.066802017390728, 0.0352981872856617, 0.001084619085304439, 0.007125324569642544, 0.001044552307575941, -0.02646971121430397, -0.06635349243879318, 0.024578111246228218, -0.012394114397466183, -0.06872833520174026, -0.00786...
Filtering with Abstract Particles
https://proceedings.mlr.press/v32/steinhardt14.html
[ "Jacob Steinhardt", "Percy Liang" ]
null
null
Using particles, beam search and sequential Monte Carlo can approximate distributions in an extremely flexible manner. However, they can suffer from sparsity and inadequate coverage on large state spaces. We present a new filtering method that addresses this issue by using “abstract particles” that each represent an e...
[]
null
82
null
null
[ -0.002690059132874012, 0.024824954569339752, 0.01474281307309866, 0.0310182124376297, 0.04432244598865509, 0.0031303942669183016, 0.00977625697851181, -0.0003116621810477227, -0.044607050716876984, -0.0592481829226017, -0.0005742043722420931, -0.005085031967610121, -0.0613827109336853, 0.0...
Stochastic Dual Coordinate Ascent with Alternating Direction Method of Multipliers
https://proceedings.mlr.press/v32/suzuki14.html
[ "Taiji Suzuki" ]
null
null
We propose a new stochastic dual coordinate ascent technique that can be applied to a wide range of regularized learning problems. Our method is based on alternating direction method of multipliers (ADMM) to deal with complex regularization functions such as structured regularizations. Although the original ADMM is a...
[]
null
83
1311.0622
title_judge
[ -0.016949670389294624, -0.007297926116734743, 0.0026023772079497576, 0.0016947367694228888, 0.029404308646917343, 0.0628267228603363, 0.013590790331363678, -0.015673266723752022, -0.03929368779063225, -0.0483785979449749, -0.004087494220584631, -0.004944708198308945, -0.040483567863702774, ...
Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction
https://proceedings.mlr.press/v32/zhou14.html
[ "Jian Zhou", "Olga Troyanskaya" ]
null
null
Predicting protein secondary structure is a fundamental problem in protein structure prediction. Here we present a new supervised generative stochastic network (GSN) based method to predict local secondary structure with deep hierarchical representations. GSN is a recently proposed deep learning technique (Bengio & Thi...
[]
null
84
1403.1347
title_snapshot
[ -0.008309374563395977, -0.021581748500466347, -0.01761108823120594, 0.02678377367556095, 0.02372918836772442, 0.009155301377177238, 0.02717275731265545, 0.002120077144354582, 0.004350560251623392, -0.03739122301340103, 0.020777011290192604, -0.008369402959942818, -0.06350929290056229, 0.03...
An Efficient Approach for Assessing Hyperparameter Importance
https://proceedings.mlr.press/v32/hutter14.html
[ "Frank Hutter", "Holger Hoos", "Kevin Leyton-Brown" ]
null
null
The performance of many machine learning methods depends critically on hyperparameter settings. Sophisticated Bayesian optimization methods have recently achieved considerable successes in optimizing these hyperparameters, in several cases surpassing the performance of human experts. However, blind reliance on such met...
[]
null
85
null
null
[ -0.007909457199275494, -0.006587993819266558, 0.006307692267000675, 0.04761195555329323, 0.018658455461263657, 0.04386164993047714, 0.052396368235349655, -0.04068434238433838, 0.004057516343891621, -0.05286049470305443, -0.00033399707172065973, 0.0046562920324504375, -0.049261026084423065, ...
An Information Geometry of Statistical Manifold Learning
https://proceedings.mlr.press/v32/suna14.html
[ "Ke Sun", "Stéphane Marchand-Maillet" ]
null
null
Manifold learning seeks low-dimensional representations of high-dimensional data. The main tactics have been exploring the geometry in an input data space and an output embedding space. We develop a manifold learning theory in a hypothesis space consisting of models. A model means a specific instance of a collection of...
[]
null
86
null
null
[ -0.031139686703681946, -0.012797903269529343, 0.00826723501086235, 0.01720559410750866, 0.029965335503220558, 0.0350174717605114, 0.016192175447940826, -0.009176085703074932, -0.026765352115035057, -0.047522563487291336, -0.024782421067357063, -0.010653151199221611, -0.06862974166870117, 0...
Relative Upper Confidence Bound for the K-Armed Dueling Bandit Problem
https://proceedings.mlr.press/v32/zoghi14.html
[ "Masrour Zoghi", "Shimon Whiteson", "Remi Munos", "Maarten Rijke" ]
null
null
This paper proposes a new method for the K-armed dueling bandit problem, a variation on the regular K-armed bandit problem that offers only relative feedback about pairs of arms. Our approach extends the Upper Confidence Bound algorithm to the relative setting by using estimates of the pairwise probabilities to select ...
[]
null
87
1312.3393
title_snapshot
[ -0.042433835566043854, -0.0009937349241226912, 0.006319268606603146, 0.06293909251689911, 0.021080531179904938, -0.0018701524240896106, 0.04526270180940628, 0.009070483036339283, -0.011977750808000565, -0.030948830768465996, -0.011346121318638325, 0.014297772198915482, -0.0635485127568245, ...
Compact Random Feature Maps
https://proceedings.mlr.press/v32/hamid14.html
[ "Raffay Hamid", "Ying Xiao", "Alex Gittens", "Dennis Decoste" ]
null
null
Kernel approximation using randomized feature maps has recently gained a lot of interest. In this work, we identify that previous approaches for polynomial kernel approximation create maps that are rank deficient, and therefore do not utilize the capacity of the projected feature space effectively. To address this chal...
[]
null
88
1312.4626
title_snapshot
[ 0.003059735521674156, -0.03787960112094879, 0.007930124178528786, 0.06975925713777542, 0.027819130569696426, 0.06557948887348175, 0.010374537669122219, -0.025170782580971718, -0.03381779044866562, -0.038464050740003586, -0.05646489933133125, -0.029100196436047554, -0.05862371623516083, 0.0...
Concentration in unbounded metric spaces and algorithmic stability
https://proceedings.mlr.press/v32/kontorovicha14.html
[ "Aryeh Kontorovich" ]
null
null
We prove an extension of McDiarmid’s inequality for metric spaces with unbounded diameter. To this end, we introduce the notion of the \em subgaussian diameter, which is a distribution-dependent refinement of the metric diameter. Our technique provides an alternative approach to that of Kutin and Niyogi’s method o...
[]
null
89
1309.1007
title_snapshot
[ -0.01872328482568264, -0.00609396630898118, 0.008574937470257282, 0.018306579440832138, 0.048120010644197464, 0.022101102396845818, 0.02125355415046215, -0.0012255636975169182, -0.027805805206298828, -0.03275858610868454, -0.022745691239833832, -0.03550773113965988, -0.09178196638822556, 0...
Heavy-tailed regression with a generalized median-of-means
https://proceedings.mlr.press/v32/hsu14.html
[ "Daniel Hsu", "Sivan Sabato" ]
null
null
This work proposes a simple and computationally efficient estimator for linear regression, and other smooth and strongly convex loss minimization problems. We prove loss approximation guarantees that hold for general distributions, including those with heavy tails. All prior results only hold for estimators which ...
[]
null
90
null
null
[ -0.010912417434155941, -0.013058900833129883, 0.013208218850195408, 0.012400020845234394, 0.042207665741443634, 0.05041484534740448, 0.019918808713555336, -0.009614347480237484, -0.032549936324357986, -0.04281359538435936, -0.025446930900216103, -0.028243454173207283, -0.06600300222635269, ...
Spectral Bandits for Smooth Graph Functions
https://proceedings.mlr.press/v32/valko14.html
[ "Michal Valko", "Remi Munos", "Branislav Kveton", "Tomáš Kocák" ]
null
null
Smooth functions on graphs have wide applications in manifold and semi-supervised learning. In this paper, we study a bandit problem where the payoffs of arms are smooth on a graph. This framework is suitable for solving online learning problems that involve graphs, such as content-based recommendation. In this problem...
[]
null
91
2604.18420
title_snapshot
[ -0.005325967445969582, -0.04042183607816696, 0.016483856365084648, 0.04881354048848152, 0.024345241487026215, 0.02580432966351509, 0.040053840726614, -0.009998822584748268, -0.005325588863343, -0.051439546048641205, -0.009029662236571312, 0.005372123327106237, -0.07359776645898819, -0.0127...
Robust Principal Component Analysis with Complex Noise
https://proceedings.mlr.press/v32/zhao14.html
[ "Qian Zhao", "Deyu Meng", "Zongben Xu", "Wangmeng Zuo", "Lei Zhang" ]
null
null
The research on robust principal component analysis (RPCA) has been attracting much attention recently. The original RPCA model assumes sparse noise, and use the L_1-norm to characterize the error term. In practice, however, the noise is much more complex and it is not appropriate to simply use a certain L_p-norm for n...
[]
null
92
null
null
[ 0.01669136993587017, 0.014275312423706055, 0.014949517324566841, 0.02773871459066868, 0.03984716907143593, 0.05432011932134628, 0.018602294847369194, -0.00108578079380095, -0.03625756874680519, -0.041705451905727386, -0.021177828311920166, -0.003723722416907549, -0.07427166402339935, 0.011...
Scalable Semidefinite Relaxation for Maximum A Posterior Estimation
https://proceedings.mlr.press/v32/huang14.html
[ "Qixing Huang", "Yuxin Chen", "Leonidas Guibas" ]
null
null
Maximum a posteriori (MAP) inference over discrete Markov random fields is a central task spanning a wide spectrum of real-world applications but known to be NP-hard for general graphs. In this paper, we propose a novel semidefinite relaxation formulation (referred to as SDR) to estimate the MAP assignment. Algorithmic...
[]
null
93
1405.4807
title_snapshot
[ -0.03829551115632057, -0.022596651688218117, -0.02181163802742958, 0.046845968812704086, 0.018078474327921867, 0.04300225153565407, 0.02326463721692562, -0.032604288309812546, -0.003929575439542532, -0.06482970714569092, 0.004889709874987602, -0.03221461921930313, -0.061075057834386826, 0....
Square Deal: Lower Bounds and Improved Relaxations for Tensor Recovery
https://proceedings.mlr.press/v32/mu14.html
[ "Cun Mu", "Bo Huang", "John Wright", "Donald Goldfarb" ]
null
null
Recovering a low-rank tensor from incomplete information is a recurring problem in signal processing and machine learning. The most popular convex relaxation of this problem minimizes the sum of the nuclear norms (SNN) of the unfolding matrices of the tensor. We show that this approach can be substantially suboptimal: ...
[]
null
94
1307.5870
title_snapshot
[ -0.04194708541035652, -0.022859007120132446, 0.013449843041598797, 0.02594713494181633, 0.010699938982725143, 0.006496736314147711, 0.0056955753825604916, -0.014399887062609196, -0.04161447286605835, -0.04350190982222557, -0.021601472049951553, 0.010435954667627811, -0.04050666093826294, -...
Automated inference of point of view from user interactions in collective intelligence venues
https://proceedings.mlr.press/v32/das14.html
[ "Sanmay Das", "Allen Lavoie" ]
null
null
Empirical evaluation of trust and manipulation in large-scale collective intelligence processes is challenging. The datasets involved are too large for thorough manual study, and current automated options are limited. We introduce a statistical framework which classifies point of view based on user interactions. The fr...
[]
null
95
null
null
[ 0.004169603809714317, -0.024641864001750946, -0.011032264679670334, 0.03353971987962723, 0.022886836901307106, -0.0027278822381049395, 0.018905390053987503, 0.014230821281671524, -0.0011109606130048633, -0.005428406875580549, -0.02052963338792324, 0.019809670746326447, -0.08087195456027985, ...
Rank-One Matrix Pursuit for Matrix Completion
https://proceedings.mlr.press/v32/wanga14.html
[ "Zheng Wang", "Ming-Jun Lai", "Zhaosong Lu", "Wei Fan", "Hasan Davulcu", "Jieping Ye" ]
null
null
Low rank matrix completion has been applied successfully in a wide range of machine learning applications, such as collaborative filtering, image inpainting and Microarray data imputation. However, many existing algorithms are not scalable to large-scale problems, as they involve computing singular value decomposition....
[]
null
96
1404.1377
title_judge
[ -0.02680782973766327, -0.02728116698563099, 0.05181551352143288, 0.01201285794377327, 0.04259863123297691, 0.02493940107524395, 0.02170131728053093, 0.0011821553343906999, -0.05905556306242943, -0.06215057522058487, -0.028588339686393738, 0.00947499554604292, -0.048115331679582596, -0.0115...
Near-Optimal Joint Object Matching via Convex Relaxation
https://proceedings.mlr.press/v32/chend14.html
[ "Yuxin Chen", "Leonidas Guibas", "Qixing Huang" ]
null
null
Joint object matching aims at aggregating information from a large collection of similar instances (e.g. images, graphs, shapes) to improve the correspondences computed between pairs of objects, typically by exploiting global map compatibility. Despite some practical advances on this problem, from the theoretical point...
[]
null
97
1402.1473
title_snapshot
[ -0.017179666087031364, -0.009999825619161129, -0.01383549626916647, 0.05274922773241997, 0.032815128564834595, 0.0591903030872345, -0.004037768580019474, 0.006988969165831804, -0.0416104793548584, -0.06704043596982956, -0.06344614177942276, -0.02285248227417469, -0.08008375763893127, -0.02...
Convex Total Least Squares
https://proceedings.mlr.press/v32/malioutov14.html
[ "Dmitry Malioutov", "Nikolai Slavov" ]
null
null
We study the total least squares (TLS) problem that generalizes least squares regression by allowing measurement errors in both dependent and independent variables. TLS is widely used in applied fields including computer vision, system identification and econometrics. The special case when all dependent and independen...
[]
null
98
1406.0189
title_snapshot
[ 0.009058119729161263, 0.009209752082824707, -0.015970706939697266, 0.020658185705542564, 0.05081132799386978, 0.02865428850054741, 0.0449952632188797, 0.016513867303729057, -0.04155772551894188, -0.02765708602964878, -0.00869676936417818, -0.029861995950341225, -0.05491770803928375, 0.0029...
On p-norm Path Following in Multiple Kernel Learning for Non-linear Feature Selection
https://proceedings.mlr.press/v32/jawanpuria14.html
[ "Pratik Jawanpuria", "Manik Varma", "Saketha Nath" ]
null
null
Our objective is to develop formulations and algorithms for efficiently computing the feature selection path – i.e. the variation in classification accuracy as the fraction of selected features is varied from null to unity. Multiple Kernel Learning subject to l_p\geq1 regularization (l_p-MKL) has been demonstrated to b...
[]
null
99
null
null
[ -0.021203897893428802, -0.009642343968153, 0.02076747640967369, 0.027624040842056274, 0.053685422986745834, 0.05402892455458641, 0.015209674835205078, -0.008874813094735146, -0.032663170248270035, -0.03674786537885666, -0.020051511004567146, 0.03204794228076935, -0.05464544892311096, -0.00...
Gradient Hard Thresholding Pursuit for Sparsity-Constrained Optimization
https://proceedings.mlr.press/v32/yuan14.html
[ "Xiaotong Yuan", "Ping Li", "Tong Zhang" ]
null
null
Hard Thresholding Pursuit (HTP) is an iterative greedy selection procedure for finding sparse solutions of underdetermined linear systems. This method has been shown to have strong theoretical guarantees and impressive numerical performance. In this paper, we generalize HTP from compressed sensing to a generic problem ...
[]
null
100
1311.5750
title_snapshot
[ -0.01486067846417427, 0.0026444564573466778, 0.009569869376718998, 0.02265036478638649, 0.029445037245750427, 0.02864094264805317, 0.019416488707065582, -0.01607425883412361, -0.05928894132375717, -0.05551567301154137, -0.01202809065580368, 0.00007071933941915631, -0.052354130893945694, -0...