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
An Optimal Policy for Target Localization with Application to Electron Microscopy
https://proceedings.mlr.press/v28/sznitman13.html
[ "Raphael Sznitman", "Aurelien Lucchi", "Peter Frazier", "Bruno Jedynak", "Pascal Fua" ]
null
null
This paper considers the task of finding a target location by making a limited number of sequential observations. Each observation results from evaluating an imperfect classifier of a chosen cost and accuracy on an interval of chosen length and position. Within a Bayesian framework, we study the problem of minimizing...
[]
null
1
null
null
[ -0.02799111045897007, 0.029664434492588043, -0.01900452934205532, 0.042116761207580566, 0.040250129997730255, 0.026807699352502823, 0.014842191711068153, -0.0025984090752899647, -0.026088017970323563, -0.05803614482283592, 0.012665338814258575, 0.005030046217143536, -0.056431181728839874, ...
Domain Generalization via Invariant Feature Representation
https://proceedings.mlr.press/v28/muandet13.html
[ "Krikamol Muandet", "David Balduzzi", "Bernhard Schölkopf" ]
null
null
This paper investigates domain generalization: How to take knowledge acquired from an arbitrary number of related domains and apply it to previously unseen domains? We propose Domain-Invariant Component Analysis (DICA), a kernel-based optimization algorithm that learns an invariant transformation by minimizing the diss...
[]
null
2
1301.2115
title_snapshot
[ -0.001415218343026936, -0.01400730200111866, 0.028742410242557526, 0.057125773280858994, 0.043010760098695755, 0.01751641556620598, 0.042020298540592194, -0.024428637698292732, -0.00825301930308342, -0.006824067793786526, -0.0334775373339653, -0.004415446426719427, -0.08466076850891113, 0....
A Spectral Learning Approach to Range-Only SLAM
https://proceedings.mlr.press/v28/boots13.html
[ "Byron Boots", "Geoff Gordon" ]
null
null
We present a novel spectral learning algorithm for simultaneous localization and mapping (SLAM) from range data with known correspondences. This algorithm is an instance of a general spectral system identification framework, from which it inherits several desirable properties, including statistical consistency and no ...
[]
null
3
1207.2491
title_snapshot
[ -0.003630173858255148, 0.004466777201741934, 0.003773541422560811, 0.017547326162457466, 0.04190154746174812, 0.03687678277492523, 0.053637370467185974, 0.018590813502669334, -0.04298128932714462, -0.05962078645825386, -0.018233157694339752, 0.009164768271148205, -0.07787549495697021, -0.0...
Near-Optimal Bounds for Cross-Validation via Loss Stability
https://proceedings.mlr.press/v28/kumar13a.html
[ "Ravi Kumar", "Daniel Lokshtanov", "Sergei Vassilvitskii", "Andrea Vattani" ]
null
null
Multi-fold cross-validation is an established practice to estimate the error rate of a learning algorithm. Quantifying the variance reduction gains due to cross-validation has been challenging due to the inherent correlations introduced by the folds. In this work we introduce a new and weak measure of stability...
[]
null
4
null
null
[ -0.01035750936716795, 0.009594383649528027, 0.0017693393165245652, 0.013246092945337296, 0.05260961875319481, 0.01196642592549324, 0.0310675036162138, -0.038437165319919586, -0.022940751165151596, -0.03848149627447128, -0.005783788859844208, -0.004003605339676142, -0.05685817822813988, -0....
Sparsity-Based Generalization Bounds for Predictive Sparse Coding
https://proceedings.mlr.press/v28/mehta13.html
[ "Nishant Mehta", "Alexander Gray" ]
null
null
The goal of predictive sparse coding is to learn a representation of examples as sparse linear combinations of elements from a dictionary, such that a learned hypothesis linear in the new representation performs well on a predictive task. Predictive sparse coding has demonstrated impressive performance on a variety of ...
[]
null
5
null
null
[ -0.018128663301467896, 0.0036546222399920225, 0.01574869081377983, 0.02877954952418804, 0.0648421198129654, 0.02819996140897274, 0.034007590264081955, 0.0054009356535971165, -0.06595014780759811, -0.02893734909594059, -0.017272667959332466, -0.0014393600868061185, -0.08511105179786682, -0....
Sparse Uncorrelated Linear Discriminant Analysis
https://proceedings.mlr.press/v28/zhang13.html
[ "Xiaowei Zhang", "Delin Chu" ]
null
null
In this paper, we develop a novel approach for sparse uncorrelated linear discriminant analysis (ULDA). Our proposal is based on characterization of all solutions of the generalized ULDA. We incorporate sparsity into the ULDA transformation by seeking the solution with minimum \ell_1-norm from all minimum dimension sol...
[]
null
6
null
null
[ -0.005618146155029535, 0.006949341390281916, 0.0053497059270739555, 0.01472395658493042, 0.04214492440223694, 0.05316932126879692, 0.01617862470448017, 0.011699269525706768, -0.01659620925784111, -0.028964204713702202, 0.003582877106964588, -0.004006312228739262, -0.09974782168865204, 0.00...
Block-Coordinate Frank-Wolfe Optimization for Structural SVMs
https://proceedings.mlr.press/v28/lacoste-julien13.html
[ "Simon Lacoste-Julien", "Martin Jaggi", "Mark Schmidt", "Patrick Pletscher" ]
null
null
We propose a randomized block-coordinate variant of the classic Frank-Wolfe algorithm for convex optimization with block-separable constraints. Despite its lower iteration cost, we show that it achieves a similar convergence rate in duality gap as the full Frank-Wolfe algorithm. We also show that, when applied to the d...
[]
null
7
1207.4747
title_snapshot
[ -0.01438090018928051, 0.00576937897130847, 0.02129337377846241, 0.03576647862792015, 0.03897221386432648, 0.03563191741704941, 0.03857488930225372, -0.009686768986284733, 0.00016992900054901838, -0.044277165085077286, -0.002165119396522641, -0.015603331848978996, -0.05153727903962135, -0.0...
Fast Probabilistic Optimization from Noisy Gradients
https://proceedings.mlr.press/v28/hennig13.html
[ "Philipp Hennig" ]
null
null
Stochastic gradient descent remains popular in large-scale machine learning, on account of its very low computational cost and robustness to noise. However, gradient descent is only linearly efficient and not transformation invariant. Scaling by a local measure can substantially improve its performance. One natural cho...
[]
null
8
null
null
[ -0.02497756853699684, 0.005059353541582823, 0.009875303134322166, 0.01469638291746378, 0.0162891186773777, 0.05795406550168991, 0.02423601970076561, 0.012937474064528942, -0.027486903592944145, -0.04407932609319687, -0.01563820242881775, -0.008594975806772709, -0.06490832567214966, -0.0010...
Stochastic Gradient Descent for Non-smooth Optimization: Convergence Results and Optimal Averaging Schemes
https://proceedings.mlr.press/v28/shamir13.html
[ "Ohad Shamir", "Tong Zhang" ]
null
null
Stochastic Gradient Descent (SGD) is one of the simplest and most popular stochastic optimization methods. While it has already been theoretically studied for decades, the classical analysis usually required non-trivial smoothness assumptions, which do not apply to many modern applications of SGD with non-smooth object...
[]
null
9
1212.1824
title_snapshot
[ -0.039619408547878265, -0.018655505031347275, 0.013472708873450756, 0.03363189473748207, 0.03198010474443436, 0.04253866523504257, 0.04777945205569267, -0.004390828777104616, -0.020290987566113472, -0.05145234614610672, -0.012383841909468174, -0.04132948815822601, -0.0505506694316864, -0.0...
Stochastic Alternating Direction Method of Multipliers
https://proceedings.mlr.press/v28/ouyang13.html
[ "Hua Ouyang", "Niao He", "Long Tran", "Alexander Gray" ]
null
null
The Alternating Direction Method of Multipliers (ADMM) has received lots of attention recently due to the tremendous demand from large-scale and data-distributed machine learning applications. In this paper, we present a stochastic setting for optimization problems with non-smooth composite objective functions. To solv...
[]
null
10
null
null
[ -0.032335828989744186, -0.004463416524231434, 0.016642143949866295, 0.0011601104633882642, 0.0035787513479590416, 0.06408865004777908, 0.04398860037326813, 0.003946315497159958, -0.04886617138981819, -0.04274648800492287, -0.017634358257055283, -0.02048543654382229, -0.04584910348057747, -...
Noisy Sparse Subspace Clustering
https://proceedings.mlr.press/v28/wang13.html
[ "Yu-Xiang Wang", "Huan Xu" ]
null
null
This paper considers the problem of subspace clustering under noise. Specifically, we study the behavior of Sparse Subspace Clustering (SSC) when either adversarial or random noise is added to the unlabelled input data points, which are assumed to lie in a union of low-dimensional subspaces. We show that a modified ve...
[]
null
11
1309.1233
title_snapshot
[ -0.0061968108639121056, -0.01391487568616867, 0.02285800874233246, 0.032186027616262436, 0.031130002811551094, 0.03735394775867462, 0.02524990774691105, -0.0038479699287563562, -0.029371412470936775, -0.03879047930240631, -0.028683792799711227, -0.028650440275669098, -0.07972270995378494, ...
Parallel Markov Chain Monte Carlo for Nonparametric Mixture Models
https://proceedings.mlr.press/v28/williamson13.html
[ "Sinead Williamson", "Avinava Dubey", "Eric Xing" ]
null
null
Nonparametric mixture models based on the Dirichlet process are an elegant alternative to finite models when the number of underlying components is unknown, but inference in such models can be slow. Existing attempts to parallelize inference in such models have relied on introducing approximations, which can lead to in...
[]
null
12
null
null
[ -0.023157255724072456, 0.009725804440677166, -0.022652821615338326, 0.04212449491024017, 0.014712701551616192, 0.05909258872270584, 0.03433898836374283, 0.005226849112659693, -0.028695646673440933, -0.050284262746572495, 0.006606575567275286, 0.014954755082726479, -0.061985768377780914, 0....
Risk Bounds and Learning Algorithms for the Regression Approach to Structured Output Prediction
https://proceedings.mlr.press/v28/giguere13.html
[ "Sébastien Giguère", "François Laviolette", "Mario Marchand", "Khadidja Sylla" ]
null
null
We provide rigorous guarantees for the regression approach to structured output prediction. We show that the quadratic regression loss is a convex surrogate of the prediction loss when the output kernel satisfies some condition with respect to the prediction loss. We provide two upper bounds of the prediction risk that...
[]
null
13
null
null
[ -0.011258676648139954, -0.014844785444438457, 0.006294021382927895, 0.021232500672340393, 0.054173793643713, 0.039672963321208954, 0.006768867839127779, -0.025230543687939644, -0.025256088003516197, -0.009221860207617283, -0.027220766991376877, 0.03503302484750748, -0.06253645569086075, 0....
Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures
https://proceedings.mlr.press/v28/bergstra13.html
[ "James Bergstra", "Daniel Yamins", "David Cox" ]
null
null
Many computer vision algorithms depend on configuration settings that are typically hand-tuned in the course of evaluating the algorithm for a particular data set. While such parameter tuning is often presented as being incidental to the algorithm, correctly setting these parameter choices is frequently critical to rea...
[]
null
14
null
null
[ -0.013773627579212189, 0.01702694222331047, -0.0032238075509667397, 0.0345228835940361, 0.03644564375281334, 0.05219649896025658, 0.022184710949659348, -0.015276207588613033, -0.0007245042943395674, -0.050248321145772934, -0.016821615397930145, 0.007838770747184753, -0.06176169216632843, -...
Gibbs Max-Margin Topic Models with Fast Sampling Algorithms
https://proceedings.mlr.press/v28/zhu13.html
[ "Jun Zhu", "Ning Chen", "Hugh Perkins", "Bo Zhang" ]
null
null
Existing max-margin supervised topic models rely on an iterative procedure to solve multiple latent SVM subproblems with additional mean-field assumptions on the desired posterior distributions. This paper presents Gibbs max-margin supervised topic models by minimizing an expected margin loss, an upper bound of the exi...
[]
null
15
null
null
[ -0.008458399213850498, -0.03879033774137497, 0.000885743647813797, 0.05858682841062546, 0.031135141849517822, -0.0068229688331484795, 0.020112765952944756, -0.008575492538511753, -0.010697108693420887, -0.031732529401779175, -0.007335814647376537, 0.01852511428296566, -0.05020507797598839, ...
Cost-Sensitive Tree of Classifiers
https://proceedings.mlr.press/v28/xu13.html
[ "Zhixiang Xu", "Matt Kusner", "Kilian Weinberger", "Minmin Chen" ]
null
null
Recently, machine learning algorithms have successfully entered large-scale real-world industrial applications (e.g. search engines and email spam filters). Here, the CPU cost during test-time must be budgeted and accounted for. In this paper, we address the challenge of balancing test-time cost and the classifier accu...
[]
null
16
1210.2771
title_snapshot
[ -0.018662216141819954, -0.002459063194692135, 0.0008155499817803502, 0.030971327796578407, 0.059275124222040176, 0.058002591133117676, 0.028702879324555397, -0.011857571080327034, 0.018118316307663918, -0.019380798563361168, 0.007935121655464172, -0.009420763701200485, -0.07108504325151443, ...
Learning Hash Functions Using Column Generation
https://proceedings.mlr.press/v28/li13a.html
[ "Xi Li", "Guosheng Lin", "Chunhua Shen", "Anton Hengel", "Anthony Dick" ]
null
null
Fast nearest neighbor searching is becoming an increasingly important tool in solving many large-scale problems. Recently a number of approaches to learning data-dependent hash functions have been developed. In this work, we propose a column generation based method for learning data-dependent hash functions on t...
[]
null
17
1303.0339
title_snapshot
[ -0.01474195159971714, 0.006973721086978912, -0.00503087742254138, 0.07685771584510803, 0.0514654777944088, 0.016007060185074806, 0.02154357358813286, -0.015328881330788136, -0.009876520372927189, -0.01743350364267826, 0.0016868725651875138, -0.029189644381403923, -0.06630818545818329, 0.01...
Combinatorial Multi-Armed Bandit: General Framework and Applications
https://proceedings.mlr.press/v28/chen13a.html
[ "Wei Chen", "Yajun Wang", "Yang Yuan" ]
null
null
We define a general framework for a large class of combinatorial multi-armed bandit (CMAB) problems, where simple arms with unknown istributions form \em super arms. In each round, a super arm is played and the outcomes of its related simple arms are observed, which helps the selection of super arms in future rounds. ...
[]
null
18
null
null
[ -0.01721523329615593, -0.02507312223315239, 0.012841788120567799, 0.034197788685560226, 0.03311409801244736, 0.023278994485735893, 0.013382921926677227, 0.0026637224946171045, -0.030772455036640167, -0.04031138867139816, -0.04225265979766846, 0.005450009368360043, -0.060257438570261, -0.03...
Near-optimal Batch Mode Active Learning and Adaptive Submodular Optimization
https://proceedings.mlr.press/v28/chen13b.html
[ "Yuxin Chen", "Andreas Krause" ]
null
null
Active learning can lead to a dramatic reduction in labeling effort. However, in many practical implementations (such as crowdsourcing, surveys, high-throughput experimental design), it is preferable to query labels for batches of examples to be labelled in parallel. While several heuristics have been proposed for bat...
[]
null
19
null
null
[ -0.025372672826051712, -0.046025827527046204, 0.00084213592344895, 0.025463087484240532, 0.014786421321332455, 0.03265176713466644, 0.020615600049495697, 0.003161785891279578, -0.02719132788479328, -0.024862559512257576, -0.013025644235312939, -0.0015755572821944952, -0.060698527842760086, ...
Convex formulations of radius-margin based Support Vector Machines
https://proceedings.mlr.press/v28/do13.html
[ "Huyen Do", "Alexandros Kalousis" ]
null
null
We consider Support Vector Machines (SVMs) learned together with linear transformations of the feature spaces on which they are applied. Under this scenario the radius of the smallest data enclosing sphere is no longer fixed. Therefore optimizing the SVM error bound by considering both the radius and the margin has the...
[]
null
20
null
null
[ -0.016751183196902275, 0.0009529826347716153, 0.03828386217355728, 0.025832008570432663, 0.03464389592409134, 0.03261470049619675, 0.028396369889378548, -0.02808556519448757, -0.035072844475507736, -0.03314239904284477, -0.01534335408359766, 0.01950056292116642, -0.0627412348985672, -0.002...
Modelling Sparse Dynamical Systems with Compressed Predictive State Representations
https://proceedings.mlr.press/v28/hamilton13.html
[ "William L. Hamilton", "Mahdi Milani Fard", "Joelle Pineau" ]
null
null
Efficiently learning accurate models of dynamical systems is of central importance for developing rational agents that can succeed in a wide range of challenging domains. The difficulty of this learning problem is particularly acute in settings with large observation spaces and partial observability. We present a new a...
[]
null
21
null
null
[ -0.0438741110265255, -0.028337180614471436, -0.018050076439976692, 0.04670863226056099, 0.043590910732746124, 0.031006883829832077, 0.006551119964569807, 0.009353652596473694, -0.04403919354081154, -0.030105365440249443, 0.01114452350884676, -0.011764935217797756, -0.08113715052604675, 0.0...
A Machine Learning Framework for Programming by Example
https://proceedings.mlr.press/v28/menon13.html
[ "Aditya Menon", "Omer Tamuz", "Sumit Gulwani", "Butler Lampson", "Adam Kalai" ]
null
null
Learning programs is a timely and interesting challenge. In Programming by Example (PBE), a system attempts to infer a program from input and output examples alone, by searching for a composition of some set of base functions. We show how machine learning can be used to speed up this seemingly hopeless search problem, ...
[]
null
22
null
null
[ -0.004910449497401714, -0.02797905169427395, -0.02310767211019993, 0.02852424420416355, 0.047775525599718094, 0.03726685419678688, 0.006727222818881273, 0.007281252648681402, -0.03829287365078926, -0.020367160439491272, -0.055365439504384995, 0.04977342486381531, -0.08548245579004288, -0.0...
Discriminatively Activated Sparselets
https://proceedings.mlr.press/v28/girshick13.html
[ "Ross Girshick", "Hyun Oh Song", "Trevor Darrell" ]
null
null
Shared representations are highly appealing due to their potential for gains in computational and statistical efficiency. Compressing a shared representation leads to greater computational savings, but at the same time can severely decrease performance on a target task. Recently, sparselets (Song et al., 2012) wer...
[]
null
23
null
null
[ 0.02756309136748314, -0.03335718810558319, -0.0012390194460749626, 0.030205965042114258, 0.030659275129437447, 0.03877829387784004, 0.002893003635108471, -0.0029350847471505404, -0.04964347183704376, -0.054317157715559006, -0.005508124828338623, -0.0034614582546055317, -0.06932765990495682, ...
The Pairwise Piecewise-Linear Embedding for Efficient Non-Linear Classification
https://proceedings.mlr.press/v28/pele13.html
[ "Ofir Pele", "Ben Taskar", "Amir Globerson", "Michael Werman" ]
null
null
Linear classiffers are much faster to learn and test than non-linear ones. On the other hand, non-linear kernels offer improved performance, albeit at the increased cost of training kernel classiffers. To use non-linear mappings with efficient linear learning algorithms, explicit embeddings that approximate popular ker...
[]
null
24
null
null
[ -0.004239935893565416, -0.03521133214235306, 0.03945333883166313, 0.03687351569533348, 0.021058715879917145, 0.046210844069719315, 0.0021787844598293304, -0.026901191100478172, -0.01819220557808876, -0.02845030464231968, -0.028561729937791824, 0.010791243053972721, -0.05516604706645012, 0....
Fixed-Point Model For Structured Labeling
https://proceedings.mlr.press/v28/li13b.html
[ "Quannan Li", "Jingdong Wang", "David Wipf", "Zhuowen Tu" ]
null
null
In this paper, we propose a simple but effective solution to the structured labeling problem: a fixed-point model. Recently, layered models with sequential classifiers/regressors have gained an increasing amount of interests for structural prediction. Here, we design an algorithm with a new perspective on layered...
[]
null
25
null
null
[ 0.00032718066358938813, -0.025912554934620857, -0.01567806489765644, 0.020357239991426468, 0.018481910228729248, 0.04279632866382599, 0.0052984291687607765, -0.020261064171791077, -0.020252058282494545, -0.01798613928258419, -0.004053136333823204, 0.0001863406359916553, -0.08012650161981583,...
Connecting the Dots with Landmarks: Discriminatively Learning Domain-Invariant Features for Unsupervised Domain Adaptation
https://proceedings.mlr.press/v28/gong13.html
[ "Boqing Gong", "Kristen Grauman", "Fei Sha" ]
null
null
Learning domain-invariant features is of vital importance to unsupervised domain adaptation, where classifiers trained on the source domain need to be adapted to a different target domain for which no labeled examples are available. In this paper, we propose a novel approach for learning such features. The central idea...
[]
null
26
null
null
[ -0.026031794026494026, -0.001711788703687489, 0.0123642198741436, 0.02065601944923401, 0.03850780427455902, 0.026567889377474785, 0.013829520903527737, -0.008247386664152145, -0.0125899538397789, -0.016768967732787132, -0.054229773581027985, 0.023675762116909027, -0.09299804270267487, 0.01...
Fast Conical Hull Algorithms for Near-separable Non-negative Matrix Factorization
https://proceedings.mlr.press/v28/kumar13b.html
[ "Abhishek Kumar", "Vikas Sindhwani", "Prabhanjan Kambadur" ]
null
null
The separability assumption (Arora et al., 2012; Donoho & Stodden, 2003) turns non-negative matrix factorization (NMF) into a tractable problem. Recently, a new class of provably-correct NMF algorithms have emerged under this assumption. In this paper, we reformulate the separable NMF problem as that of finding the ext...
[]
null
27
1210.1190
title_snapshot
[ -0.03785103186964989, -0.033328086137771606, 0.025560511276125908, -0.0067296880297362804, 0.03080003149807453, 0.05151522904634476, 0.00013229160686023533, -0.016814330592751503, -0.03120291233062744, -0.06121218949556351, 0.004939166363328695, 0.005831468850374222, -0.06274031847715378, ...
Principal Component Analysis on non-Gaussian Dependent Data
https://proceedings.mlr.press/v28/han13.html
[ "Fang Han", "Han Liu" ]
null
null
In this paper, we analyze the performance of a semiparametric principal component analysis named Copula Component Analysis (COCA) (Han & Liu, 2012) when the data are dependent. The semiparametric model assumes that, after unspecified marginally monotone transformations, the distributions are multivariate Gaussian. We s...
[]
null
28
null
null
[ -0.01310700736939907, -0.02351241186261177, 0.005211386829614639, 0.010684030130505562, 0.04456622898578644, 0.06042120233178139, 0.0068926322273910046, -0.00015592516865581274, -0.016851119697093964, -0.05033855512738228, -0.002305530244484544, 0.00627902103587985, -0.08160418272018433, 0...
Learning Linear Bayesian Networks with Latent Variables
https://proceedings.mlr.press/v28/anandkumar13.html
[ "Animashree Anandkumar", "Daniel Hsu", "Adel Javanmard", "Sham Kakade" ]
null
null
This work considers the problem of learning linear Bayesian networks when some of the variables are unobserved. Identifiability and efficient recovery from low-order observable moments are established under a novel graphical constraint. The constraint concerns the expansion properties of the underlying directed ac...
[]
null
29
null
null
[ -0.003771007526665926, 0.027260998263955116, -0.024792231619358063, 0.032348357141017914, 0.04997852072119713, 0.02871602028608322, 0.0366705097258091, 0.004076430108398199, -0.01350671611726284, -0.03060346655547619, 0.02087557315826416, 0.0048454697243869305, -0.07762245088815689, 0.0013...
Multiple Identifications in Multi-Armed Bandits
https://proceedings.mlr.press/v28/bubeck13.html
[ "Séebastian Bubeck", "Tengyao Wang", "Nitin Viswanathan" ]
null
null
We study the problem of identifying the top m arms in a multi-armed bandit game. Our proposed solution relies on a new algorithm based on successive rejects of the seemingly bad arms, and successive accepts of the good ones. This algorithmic contribution allows to tackle other multiple identifications settings that wer...
[]
null
30
1205.3181
title_snapshot
[ -0.04197639971971512, 0.003330617444589734, -0.012209494598209858, 0.030262911692261696, 0.04650326818227768, 0.02959725819528103, 0.02384040504693985, -0.0037091299891471863, -0.0345226414501667, -0.045257508754730225, -0.017290586605668068, 0.0076683284714818, -0.05631394311785698, -0.02...
Learning Optimally Sparse Support Vector Machines
https://proceedings.mlr.press/v28/cotter13.html
[ "Andrew Cotter", "Shai Shalev-Shwartz", "Nati Srebro" ]
null
null
We show how to train SVMs with an optimal guarantee on the number of support vectors (up to constants), and with sample complexity and training runtime bounds matching the best known for kernel SVM optimization (i.e. without any additional asymptotic cost beyond standard SVM training). Our method is simple to implement...
[]
null
31
null
null
[ -0.008935348130762577, -0.029009267687797546, 0.04816151782870293, 0.03732690587639809, 0.043625619262456894, 0.047272998839616776, 0.025212131440639496, -0.016676578670740128, -0.05053947865962982, -0.026263905689120293, -0.03166715055704117, 0.04039657860994339, -0.054504722356796265, 0....
Dynamic Probabilistic Models for Latent Feature Propagation in Social Networks
https://proceedings.mlr.press/v28/heaukulani13.html
[ "Creighton Heaukulani", "Zoubin Ghahramani" ]
null
null
Current Bayesian models for dynamic social network data have focused on modelling the influence of evolving unobserved structure on observed social interactions. However, an understanding of how observed social relationships from the past affect future unobserved structure in the network has been neglected. In this pa...
[]
null
32
null
null
[ 0.01161724328994751, -0.022380366921424866, 0.007083614356815815, 0.025451183319091797, 0.05889070779085159, 0.008372832089662552, 0.04943402111530304, 0.016136448830366135, -0.026151563972234726, -0.03254952281713486, 0.00667034275829792, -0.012247372418642044, -0.049163877964019775, 0.00...
Efficient Sparse Group Feature Selection via Nonconvex Optimization
https://proceedings.mlr.press/v28/xiang13.html
[ "Shuo Xiang", "Xiaoshen Tong", "Jieping Ye" ]
null
null
Sparse feature selection has been demonstrated to be effective in handling high-dimensional data. While promising, most of the existing works use convex methods, which may be suboptimal in terms of the accuracy of feature selection and parameter estimation. In this paper, we expand a nonconvex paradigm to sparse group ...
[]
null
33
1205.5075
title_snapshot
[ -0.015522884204983711, -0.019153524190187454, 0.02061866596341133, 0.03734348714351654, 0.026856012642383575, 0.051245976239442825, 0.028516847640275955, -0.0020333584398031235, -0.020789960399270058, -0.05724881589412689, -0.017139827832579613, -0.001938814646564424, -0.07060898840427399, ...
Domain Adaptation for Sequence Labeling Tasks with a Probabilistic Language Adaptation Model
https://proceedings.mlr.press/v28/xiao13.html
[ "Min Xiao", "Yuhong Guo" ]
null
null
In this paper, we propose to address the problem of domain adaptation for sequence labeling tasks via distributed representation learning by using a log-bilinear language adaptation model. The proposed neural probabilistic language model simultaneously models two different but related data distributions in the source a...
[]
null
34
null
null
[ -0.03586633875966072, 0.0020110078621655703, -0.007713151164352894, 0.02688002400100231, 0.036313947290182114, 0.023966412991285324, 0.034173112362623215, -0.0004991691093891859, -0.008699566125869751, 0.005289206746965647, -0.03199462220072746, 0.02048237808048725, -0.05868116393685341, 0...
Maximum Variance Correction with Application to A* Search
https://proceedings.mlr.press/v28/chen13c.html
[ "Wenlin Chen", "Kilian Weinberger", "Yixin Chen" ]
null
null
In this paper we introduce Maximum Variance Correction (MVC), which finds large-scale feasible solutions to Maximum Variance Unfolding (MVU) by post-processing embeddings from any manifold learning algorithm. It increases the scale of MVU embeddings by several orders of magnitude and is naturally parallel. This unprece...
[]
null
35
null
null
[ -0.019827499985694885, 0.006849427707493305, -0.005748538766056299, 0.030394120141863823, 0.03535810112953186, 0.0193245280534029, 0.06520521640777588, 0.00706724775955081, -0.00003049234328500461, -0.07868775725364685, 0.0014328404795378447, -0.00834234245121479, -0.08022695034742355, -0....
Adaptive Sparsity in Gaussian Graphical Models
https://proceedings.mlr.press/v28/wong13.html
[ "Eleanor Wong", "Suyash Awate", "P. Thomas Fletcher" ]
null
null
An effective approach to structure learning and parameter estimation for Gaussian graphical models is to impose a sparsity prior, such as a Laplace prior, on the entries of the precision matrix. Such an approach involves a hyperparameter that must be tuned to control the amount of sparsity. In this paper, we introduce ...
[]
null
36
null
null
[ -0.01013495959341526, 0.02041780762374401, -0.0002805768745020032, 0.018856482580304146, 0.03300591558218002, 0.03099348209798336, 0.042451679706573486, -0.0059930202551186085, -0.02362155355513096, -0.06279760599136353, 0.027851471677422523, 0.006944240536540747, -0.07872944325208664, 0.0...
Average Reward Optimization Objective In Partially Observable Domains
https://proceedings.mlr.press/v28/grinberg13.html
[ "Yuri Grinberg", "Doina Precup" ]
null
null
We consider the problem of average reward optimization in domains with partial observability, within the modeling framework of linear predictive state representations (PSRs). The key to average-reward computation is to have a well-defined stationary behavior of a system, so the required averages can be computed. If, ad...
[]
null
37
null
null
[ -0.06886978447437286, -0.0007984259282238781, -0.0018293180037289858, 0.051905158907175064, 0.04487388953566551, 0.011554942466318607, 0.030440615490078926, -0.011516736820340157, -0.024131132289767265, -0.03943828493356705, -0.018662946298718452, -0.024060579016804695, -0.0873861163854599, ...
Feature Selection in High-Dimensional Classification
https://proceedings.mlr.press/v28/kolar13.html
[ "Mladen Kolar", "Han Liu" ]
null
null
High-dimensional discriminant analysis is of fundamental importance in multivariate statistics. Existing theoretical results sharply characterize different procedures, providing sharp convergence results for the classification risk, as well as the l2 convergence results to the discriminative rule. However, sharp theore...
[]
null
38
null
null
[ -0.01785845123231411, -0.00034385742037557065, 0.0002393151808064431, 0.021994346752762794, 0.040507201105356216, 0.051786016672849655, 0.03428336977958679, -0.006619847379624844, -0.006487135309726, -0.035412274301052094, -0.013726349920034409, 0.016612371429800987, -0.08998730778694153, ...
Human Boosting
https://proceedings.mlr.press/v28/pareek13.html
[ "Harsh Pareek", "Pradeep Ravikumar" ]
null
null
Humans may be exceptional learners but they have biological limitations and moreover, inductive biases similar to machine learning algorithms. This puts limits on human learning ability and on the kinds of learning tasks humans can easily handle. In this paper, we consider the problem of “boosting” human learners to ex...
[]
null
39
null
null
[ 0.017437180504202843, -0.0353725291788578, 0.002947036875411868, 0.03060644492506981, 0.032024044543504715, -0.02831518091261387, 0.02094786986708641, 0.028555331751704216, -0.013472765684127808, -0.03677305579185486, -0.01797952502965927, 0.03402099758386612, -0.07426317036151886, -0.0343...
Efficient Dimensionality Reduction for Canonical Correlation Analysis
https://proceedings.mlr.press/v28/avron13.html
[ "Haim Avron", "Christos Boutsidis", "Sivan Toledo", "Anastasios Zouzias" ]
null
null
We present a fast algorithm for approximate Canonical Correlation Analysis (CCA). Given a pair of tall-and-thin matrices, the proposed algorithm first employs a randomized dimensionality reduction transform to reduce the size of the input matrices, and then applies any standard CCA algorithm to the new pair of matrice...
[]
null
40
1209.2185
title_snapshot
[ 0.001574234222061932, -0.007418881170451641, -0.00013797840801998973, 0.022396044805645943, 0.0626111701130867, 0.03519199788570404, 0.01882285252213478, 0.006147733423858881, -0.013287202455103397, -0.02682630717754364, -0.011464421637356281, -0.02093920111656189, -0.08084216713905334, -0...
Parsing epileptic events using a Markov switching process model for correlated time series
https://proceedings.mlr.press/v28/wulsin13.html
[ "Drausin Wulsin", "Emily Fox", "Brian Litt" ]
null
null
Patients with epilepsy can manifest short, sub-clinical epileptic “bursts” in addition to full-blown clinical seizures. We believe the relationship between these two classes of events—something not previously studied quantitatively—could yield important insights into the nature and intrinsic dynamics of seizures. A goa...
[]
null
41
null
null
[ -0.025793597102165222, 0.013761989772319794, -0.019258640706539154, 0.0018922581803053617, 0.049453459680080414, -0.004163186065852642, 0.033028844743967056, 0.022396394982933998, -0.022666431963443756, -0.03595941513776779, -0.0006215940811671317, 0.015762202441692352, -0.051956906914711, ...
Optimal rates for stochastic convex optimization under Tsybakov noise condition
https://proceedings.mlr.press/v28/ramdas13.html
[ "Aaditya Ramdas", "Aarti Singh" ]
null
null
We focus on the problem of minimizing a convex function f over a convex set S given T queries to a stochastic first order oracle. We argue that the complexity of convex minimization is only determined by the rate of growth of the function around its minimum x^*_f,S, as quantified by a Tsybakov-like noise condition. Spe...
[]
null
42
1207.3012
title_judge
[ -0.029894676059484482, 0.028386663645505905, 0.011383791454136372, 0.028874320909380913, 0.028087511658668518, 0.04424309358000755, 0.02163432165980339, 0.024640223011374474, -0.007096528075635433, -0.0337984599173069, -0.029686374589800835, 0.014132224954664707, -0.055585578083992004, -0....
A Randomized Mirror Descent Algorithm for Large Scale Multiple Kernel Learning
https://proceedings.mlr.press/v28/afkanpour13.html
[ "Arash Afkanpour", "András György", "Csaba Szepesvari", "Michael Bowling" ]
null
null
We consider the problem of simultaneously learning to linearly combine a very large number of kernels and learn a good predictor based on the learnt kernel. When the number of kernels d to be combined is very large, multiple kernel learning methods whose computational cost scales linearly in d are intractable. We propo...
[]
null
43
1205.0288
title_snapshot
[ -0.0258588008582592, -0.04004748538136482, 0.029476070776581764, 0.048409271985292435, 0.025672107934951782, 0.04670632258057594, 0.0009295229101553559, -0.012555161491036415, -0.019530784338712692, -0.036318812519311905, -0.01717761717736721, 0.024733711034059525, -0.04991350695490837, -0...
Noisy and Missing Data Regression: Distribution-Oblivious Support Recovery
https://proceedings.mlr.press/v28/chen13d.html
[ "Yudong Chen", "Constantine Caramanis" ]
null
null
Many models for sparse regression typically assume that the covariates are known completely, and without noise. Particularly in high-dimensional applications, this is often not the case. Worse yet, even estimating statistics of the noise (the noise covariance) can be a central challenge. In this paper we develop a simp...
[]
null
44
null
null
[ -0.030044620856642723, 0.0023229429498314857, 0.017301779240369797, 0.01754092238843441, 0.04908842220902443, 0.05637471377849579, 0.03465329855680466, -0.005669047124683857, -0.04576723277568817, -0.044358666986227036, -0.023300034925341606, 0.008550286293029785, -0.08318568021059036, -0....
Dual Averaging and Proximal Gradient Descent for Online Alternating Direction Multiplier Method
https://proceedings.mlr.press/v28/suzuki13.html
[ "Taiji Suzuki" ]
null
null
We develop new stochastic optimization methods that are applicable to a wide range of structured regularizations. Basically our methods are combinations of basic stochastic optimization techniques and Alternating Direction Multiplier Method (ADMM). ADMM is a general framework for optimizing a composite function, ...
[]
null
45
null
null
[ -0.014818649739027023, -0.00968370120972395, 0.028676200658082962, -0.0010852875420823693, 0.03799109160900116, 0.04586626589298248, 0.0335170142352581, -0.019248830154538155, -0.0447533093392849, -0.038038820028305054, -0.0018147403607144952, -0.020388783887028694, -0.042491115629673004, ...
A New Frontier of Kernel Design for Structured Data
https://proceedings.mlr.press/v28/shin13.html
[ "Kilho Shin" ]
null
null
Many kernels for discretely structured data in the literature are designed within the framework of the convolution kernel and its generalization, the mapping kernel. The two most important advantages to use this framework is an easy-to-check criteria of positive definiteness and efficient computation based on the dynam...
[]
null
46
null
null
[ -0.016494885087013245, -0.03839355707168579, 0.01772455684840679, 0.05705295503139496, 0.034067243337631226, 0.02893691323697567, -0.00467075826600194, -0.023022174835205078, 0.0014049875317141414, -0.030790328979492188, -0.011505581438541412, -0.012191218324005604, -0.06294938176870346, 0...
Learning with Marginalized Corrupted Features
https://proceedings.mlr.press/v28/vandermaaten13.html
[ "Laurens Maaten", "Minmin Chen", "Stephen Tyree", "Kilian Weinberger" ]
null
null
The goal of machine learning is to develop predictors that generalize well to test data. Ideally, this is achieved by training on very large (infinite) training data sets that capture all variations in the data distribution. In the case of finite training data, an effective solution is to extend the training set with a...
[]
null
47
null
null
[ 0.002523200586438179, -0.02306007407605648, -0.013950598426163197, 0.07642972469329834, 0.07732407003641129, 0.04324488714337349, -0.0013585156993940473, -0.029415378347039223, -0.03515564650297165, -0.025658156722784042, -0.011056598275899887, 0.037264544516801834, -0.07064458727836609, 0...
Approximation properties of DBNs with binary hidden units and real-valued visible units
https://proceedings.mlr.press/v28/krause13.html
[ "Oswin Krause", "Asja Fischer", "Tobias Glasmachers", "Christian Igel" ]
null
null
Deep belief networks (DBNs) can approximate any distribution over fixed-length binary vectors. However, DBNs are frequently applied to model real-valued data, and so far little is known about their representational power in this case. We analyze the approximation properties of DBNs with two layers of binary hidden uni...
[]
null
48
null
null
[ -0.018056131899356842, 0.003966000396758318, -0.0053065805695950985, 0.04615374282002449, 0.040008388459682465, 0.04554304853081703, 0.028327472507953644, 0.0013127505080774426, -0.0287444107234478, -0.02895139716565609, -0.01631011627614498, 0.006493312772363424, -0.04796889051795006, 0.0...
Revisiting Frank-Wolfe: Projection-Free Sparse Convex Optimization
https://proceedings.mlr.press/v28/jaggi13.html
[ "Martin Jaggi" ]
null
null
We provide stronger and more general primal-dual convergence results for Frank-Wolfe-type algorithms (a.k.a. conditional gradient) for constrained convex optimization, enabled by a simple framework of duality gap certificates. Our analysis also holds if the linear subproblems are only solved approximately (as well as i...
[]
null
49
null
null
[ -0.025290267542004585, 0.00015419474220834672, 0.03969390317797661, 0.03826841339468956, 0.029973573982715607, 0.03730352967977524, 0.014106697402894497, 0.008188709616661072, -0.02303788810968399, -0.04790782928466797, -0.004911998752504587, 0.01053232979029417, -0.06204112991690636, -0.0...
General Functional Matrix Factorization Using Gradient Boosting
https://proceedings.mlr.press/v28/chen13e.html
[ "Tianqi Chen", "Hang Li", "Qiang Yang", "Yong Yu" ]
null
null
Matrix factorization is among the most successful techniques for collaborative filtering. One challenge of collaborative filtering is how to utilize available auxiliary information to improve prediction accuracy. In this paper, we study the problem of utilizing auxiliary information as features of factorization and ...
[]
null
50
null
null
[ 0.019153103232383728, -0.052154481410980225, 0.04730600491166115, 0.010472198948264122, 0.06700628995895386, 0.02623242884874344, 0.01882198080420494, -0.013439403846859932, -0.0010224502766504884, -0.04231730103492737, -0.015188329853117466, 0.009485378861427307, -0.07682375609874725, -0....
Iterative Learning and Denoising in Convolutional Neural Associative Memories
https://proceedings.mlr.press/v28/karbasi13.html
[ "Amin Karbasi", "Amir Hesam Salavati", "Amin Shokrollahi" ]
null
null
The task of a neural associative memory is to retrieve a set of previously memorized patterns from their noisy versions by using a network of neurons. Hence, an ideal network should be able to 1) gradually learn a set of patterns, 2) retrieve the correct pattern from noisy queries and 3) maximize the number of memorize...
[]
null
51
null
null
[ -0.019045954570174217, 0.006225986406207085, 0.007668411359190941, 0.0386626198887825, 0.01896628923714161, 0.04278699681162834, 0.011161675676703453, 0.019524063915014267, -0.03490936756134033, -0.027082130312919617, 0.0006751614855602384, 0.0018158196471631527, -0.04963346943259239, 0.00...
Scaling Multidimensional Gaussian Processes using Projected Additive Approximations
https://proceedings.mlr.press/v28/gilboa13.html
[ "Elad Gilboa", "Yunus Saatçi", "John Cunningham", "Elad Gilboa" ]
null
null
Exact Gaussian Process (GP) regression has O(N^3) runtime for data size N, making it intractable for large N. Advances in GP scaling have not been extended to the multidimensional input setting, despite the preponderance of multidimensional applications. This paper introduces and tests a novel method of projected add...
[]
null
52
1209.4120
title_judge
[ -0.010984936729073524, 0.0021671904250979424, 0.022777361795306206, -0.006678303703665733, 0.02933472767472267, 0.05866358056664467, 0.00868269894272089, -0.015089056454598904, -0.0380936861038208, -0.03889104351401329, -0.004334432538598776, 0.0020815981552004814, -0.08353491127490997, 0....
Active Learning for Multi-Objective Optimization
https://proceedings.mlr.press/v28/zuluaga13.html
[ "Marcela Zuluaga", "Guillaume Sergent", "Andreas Krause", "Markus Püschel" ]
null
null
In many fields one encounters the challenge of identifying, out of a pool of possible designs, those that simultaneously optimize multiple objectives. This means that usually there is not one optimal design but an entire set of Pareto-optimal ones with optimal tradeoffs in the objectives. In many applications, evaluati...
[]
null
53
null
null
[ -0.04812590405344963, -0.003135360311716795, 0.010474924929440022, 0.013628428801894188, 0.0240885429084301, 0.048889365047216415, -0.024381287395954132, -0.019168904051184654, -0.006041668821126223, -0.05098900943994522, -0.011099551804363728, 0.016080625355243683, -0.06630513072013855, -...
A Generalized Kernel Approach to Structured Output Learning
https://proceedings.mlr.press/v28/kadri13.html
[ "Hachem Kadri", "Mohammad Ghavamzadeh", "Philippe Preux" ]
null
null
We study the problem of structured output learning from a regression perspective. We first provide a general formulation of the kernel dependency estimation (KDE) approach to this problem using operator-valued kernels. Our formulation overcomes the two main limitations of the original KDE approach, namely the decouplin...
[]
null
54
1205.2171
title_snapshot
[ 0.01026238314807415, -0.005363663658499718, 0.03215231001377106, 0.04268861562013626, 0.041214846074581146, 0.05029020458459854, -0.006898144725710154, -0.02025515027344227, 0.002172072883695364, -0.028093162924051285, -0.043951161205768585, 0.04078098386526108, -0.0809161588549614, 0.0143...
Efficient Active Learning of Halfspaces: an Aggressive Approach
https://proceedings.mlr.press/v28/gonen13.html
[ "Alon Gonen", "Sivan Sabato", "Shai Shalev-Shwartz" ]
null
null
We study pool-based active learning of half-spaces. We revisit the aggressive approach for active learning in the realizable case, and show that it can be made efficient and practical, while also having theoretical guarantees under reasonable assumptions. We further show, both theoretically and experimentally, that it...
[]
null
55
1208.3561
title_snapshot
[ 0.004745830316096544, -0.025093713775277138, -0.0048820641823112965, 0.02216631919145584, 0.03900827467441559, 0.018835846334695816, -0.013565951958298683, -0.00919428188353777, -0.024253860116004944, -0.033580340445041656, -0.016053371131420135, 0.009917334653437138, -0.07456901669502258, ...
Enhanced statistical rankings via targeted data collection
https://proceedings.mlr.press/v28/osting13.html
[ "Braxton Osting", "Christoph Brune", "Stanley Osher" ]
null
null
Given a graph where vertices represent alternatives and pairwise comparison data, y_ij, is given on the edges, the statistical ranking problem is to find a potential function, defined on the vertices, such that the gradient of the potential function agrees with pairwise comparisons. We study the dependence of the stati...
[]
null
56
null
null
[ -0.01778673380613327, -0.03377917408943176, 0.01349195558577776, 0.05241461843252182, 0.0289397481828928, 0.027085544541478157, 0.007781951688230038, -0.04170309007167816, -0.027593279257416725, -0.06134003400802612, -0.011550769209861755, -0.005104673560708761, -0.07471009343862534, -0.00...
Online Feature Selection for Model-based Reinforcement Learning
https://proceedings.mlr.press/v28/nguyen13.html
[ "Trung Nguyen", "Zhuoru Li", "Tomi Silander", "Tze Yun Leong" ]
null
null
We propose a new framework for learning the world dynamics of feature-rich environments in model-based reinforcement learning. The main idea is formalized as a new, factored state-transition representation that supports efficient online-learning of the relevant features. We construct the transition models through predi...
[]
null
57
null
null
[ -0.0371633879840374, -0.008155454881489277, 0.004097525030374527, 0.037000130861997604, 0.05540008470416069, 0.028050126507878304, 0.0075531648471951485, 0.010795999318361282, -0.023518290370702744, -0.024426467716693878, -0.017716480419039726, 0.007021422032266855, -0.06405553221702576, -...
ELLA: An Efficient Lifelong Learning Algorithm
https://proceedings.mlr.press/v28/ruvolo13.html
[ "Paul Ruvolo", "Eric Eaton" ]
null
null
The problem of learning multiple consecutive tasks, known as lifelong learning, is of great importance to the creation of intelligent, general-purpose, and flexible machines. In this paper, we develop a method for online multi-task learning in the lifelong learning setting. The proposed Efficient Lifelong Learning Al...
[]
null
58
null
null
[ -0.009189511649310589, -0.02115803211927414, -0.022382786497473717, 0.016742980107665062, 0.030983658507466316, 0.02760494314134121, 0.0018239469500258565, -0.013976759277284145, -0.035338860005140305, -0.03595234826207161, -0.0299752838909626, 0.01152355968952179, -0.052266404032707214, -...
A Structural SVM Based Approach for Optimizing Partial AUC
https://proceedings.mlr.press/v28/narasimhan13.html
[ "Harikrishna Narasimhan", "Shivani Agarwal" ]
null
null
The area under the ROC curve (AUC) is a widely used performance measure in machine learning. Increasingly, however, in several applications, ranging from ranking and biometric screening to medical diagnosis, performance is measured not in terms of the full area under the ROC curve, but instead, in terms of the partial ...
[]
null
59
null
null
[ -0.03032175451517105, -0.0034381512086838484, -0.0152535829693079, 0.014998422004282475, 0.029386235401034355, 0.038568247109651566, 0.047289278358221054, -0.02299574390053749, -0.014278141781687737, -0.054471150040626526, -0.005859397817403078, -0.004314040299504995, -0.04652394726872444, ...
Convex Relaxations for Learning Bounded-Treewidth Decomposable Graphs
https://proceedings.mlr.press/v28/kumar13c.html
[ "K. S. Sesh Kumar", "Francis Bach" ]
null
null
We consider the problem of learning the structure of undirected graphical models with bounded treewidth, within the maximum likelihood framework. This is an NP-hard problem and most approaches consider local search techniques. In this paper, we pose it as a combinatorial optimization problem, which is then relaxed to a...
[]
null
60
1212.2573
title_snapshot
[ -0.007977218367159367, -0.0065123592503368855, -0.0031148220878094435, 0.05167660117149353, 0.035039570182561874, 0.030844440683722496, 0.034525685012340546, 0.010228725150227547, -0.002458177739754319, -0.038216590881347656, -0.00840504840016365, -0.00967899989336729, -0.07878628373146057, ...
Adaptive Task Assignment for Crowdsourced Classification
https://proceedings.mlr.press/v28/ho13.html
[ "Chien-Ju Ho", "Shahin Jabbari", "Jennifer Wortman Vaughan" ]
null
null
Crowdsourcing markets have gained popularity as a tool for inexpensively collecting data from diverse populations of workers. Classification tasks, in which workers provide labels (such as “offensive” or “not offensive”) for instances (such as websites), are among the most common tasks posted, but due to a mix of human...
[]
null
61
null
null
[ 0.009959581308066845, -0.022997722029685974, -0.030420808121562004, 0.013305142521858215, 0.011507156305015087, 0.036725256592035294, 0.0062907119281589985, 0.0008847920107655227, -0.02179623208940029, -0.023950990289449692, -0.04457416385412216, 0.010937755927443504, -0.09811092168092728, ...
Optimal Regret Bounds for Selecting the State Representation in Reinforcement Learning
https://proceedings.mlr.press/v28/maillard13.html
[ "Odalric-Ambrym Maillard", "Phuong Nguyen", "Ronald Ortner", "Daniil Ryabko" ]
null
null
We consider an agent interacting with an environment in a single stream of actions, observations, and rewards, with no reset. This process is not assumed to be a Markov Decision Process (MDP). Rather, the agent has several representations (mapping histories of past interactions to a discrete state space) of the environ...
[]
null
62
1302.2553
title_snapshot
[ -0.07274000346660614, -0.0018830744083970785, -0.0300262663513422, 0.05437951534986496, 0.04979736730456352, 0.030561039224267006, 0.022850386798381805, 0.008973146788775921, -0.01870146580040455, -0.04907449334859848, -0.028145447373390198, 0.0036901552230119705, -0.07578805834054947, -0....
Better Mixing via Deep Representations
https://proceedings.mlr.press/v28/bengio13.html
[ "Yoshua Bengio", "Gregoire Mesnil", "Yann Dauphin", "Salah Rifai" ]
null
null
It has been hypothesized, and supported with experimental evidence, that deeper representations, when well trained, tend to do a better job at disentangling the underlying factors of variation. We study the following related conjecture: better representations, in the sense of better disentangling, can be exploited to ...
[]
null
63
1207.4404
title_snapshot
[ 0.012415915727615356, 0.0006088863592594862, -0.032478734850883484, 0.04697461426258087, 0.06055842712521553, 0.011423694901168346, 0.04631512984633446, 0.009220018051564693, -0.042059510946273804, -0.059805937111377716, 0.0008376705809496343, -0.030580351129174232, -0.0842040479183197, 0....
Online Latent Dirichlet Allocation with Infinite Vocabulary
https://proceedings.mlr.press/v28/zhai13.html
[ "Ke Zhai", "Jordan Boyd-Graber" ]
null
null
Topic models based on latent Dirichlet allocation (LDA) assume a predefined vocabulary a priori. This is reasonable in batch settings, but it is not reasonable when data are revealed over time, as is the case with streaming / online algorithms. To address this lacuna, we extend LDA by drawing topics from a Dirichlet pr...
[]
null
64
null
null
[ -0.008612512610852718, -0.021070824936032295, -0.010938199236989021, 0.06383504718542099, 0.03614598512649536, 0.024307910352945328, -0.006426471751183271, 0.0190963763743639, 0.012237529270350933, -0.008416919969022274, -0.048129573464393616, 0.005756730679422617, -0.0656178891658783, 0.0...
Characterizing the Representer Theorem
https://proceedings.mlr.press/v28/yu13.html
[ "Yaoliang Yu", "Hao Cheng", "Dale Schuurmans", "Csaba Szepesvari" ]
null
null
The representer theorem assures that kernel methods retain optimality under penalized empirical risk minimization. While a sufficient condition on the form of the regularizer guaranteeing the representer theorem has been known since the initial development of kernel methods, necessary conditions have only been investig...
[]
null
65
null
null
[ -0.04691753163933754, -0.032779596745967865, 0.02991735003888607, 0.0351485051214695, 0.045728541910648346, 0.05379355326294899, 0.003079373622313142, -0.043584004044532776, -0.04031224176287651, -0.0345773883163929, 0.0011052455520257354, 0.01783052645623684, -0.038759779185056686, 0.0307...
Dynamical Models and tracking regret in online convex programming
https://proceedings.mlr.press/v28/hall13.html
[ "Eric Hall", "Rebecca Willett" ]
null
null
This paper describes a new online convex optimization method which incorporates a family of candidate dynamical models and establishes novel tracking regret bounds that scale with comparator’s deviation from the best dynamical model in this family. Previous online optimization methods are designed to have a total accum...
[]
null
66
1301.1254
title_snapshot
[ -0.0443563349545002, 0.0005182938184589148, -0.011084924452006817, 0.03767291083931923, 0.033182140439748764, 0.047329772263765335, 0.015535430051386356, 0.042734622955322266, -0.044823139905929565, -0.046221788972616196, -0.00197145389392972, -0.0018813912756741047, -0.04802973195910454, ...
Large-Scale Bandit Problems and KWIK Learning
https://proceedings.mlr.press/v28/abernethy13.html
[ "Jacob Abernethy", "Kareem Amin", "Michael Kearns", "Moez Draief" ]
null
null
We show that parametric multi-armed bandit (MAB) problems with large state and action spaces can be algorithmically reduced to the supervised learning model known as Knows What It Knows or KWIK learning. We give matching impossibility results showing that the KWIK learnability requirement cannot be replaced by weaker s...
[]
null
67
null
null
[ -0.012343530543148518, -0.015956265851855278, -0.019469933584332466, 0.03639473766088486, 0.03011247329413891, 0.03245653584599495, 0.02832362800836563, -0.0061891996301710606, -0.021905234083533287, -0.022769583389163017, -0.010496851988136768, 0.01822097972035408, -0.06957009434700012, -...
Vanishing Component Analysis
https://proceedings.mlr.press/v28/livni13.html
[ "Roi Livni", "David Lehavi", "Sagi Schein", "Hila Nachliely", "Shai Shalev-Shwartz", "Amir Globerson" ]
null
null
The vanishing ideal of a set of n points S, is the set of all polynomials that attain the value of zero on all the points in S. Such ideals can be compactly represented using a small set of polynomials known as generators of the ideal. Here we describe and analyze an efficient procedure that constructs a set of generat...
[]
null
68
null
null
[ -0.014437234029173851, 0.013713675551116467, 0.023324497044086456, 0.04678202420473099, 0.031586140394210815, 0.02290254272520542, 0.012814968824386597, -0.02576262317597866, -0.04718061164021492, -0.031131574884057045, -0.07148166745901108, 0.0022825796622782946, -0.06896620243787766, 0.0...
Learning an Internal Dynamics Model from Control Demonstration
https://proceedings.mlr.press/v28/golub13.html
[ "Matthew Golub", "Steven Chase", "Byron Yu" ]
null
null
Much work in optimal control and inverse control has assumed that the controller has perfect knowledge of plant dynamics. However, if the controller is a human or animal subject, the subject’s internal dynamics model may differ from the true plant dynamics. Here, we consider the problem of learning the subject’s inte...
[]
null
69
null
null
[ -0.05637303367257118, -0.0059900786727666855, -0.029506297782063484, 0.026217050850391388, 0.05084172636270523, 0.015759674832224846, 0.037542663514614105, 0.012965168803930283, -0.05690417438745499, -0.02497352473437786, -0.004751197993755341, 0.004543743561953306, -0.03598739951848984, -...
Robust Structural Metric Learning
https://proceedings.mlr.press/v28/lim13.html
[ "Daryl Lim", "Gert Lanckriet", "Brian McFee" ]
null
null
Metric learning algorithms produce a linear transformation of data which is optimized for a prediction task, such as nearest-neighbor classification or ranking. However, when the input data contains a large portion of non-informative features, existing methods fail to identify the relevant features, and performance de...
[]
null
70
null
null
[ 0.004680550657212734, 0.006555987987667322, 0.01784531958401203, 0.02827884815633297, 0.03312038257718086, 0.03389190137386322, 0.017795661464333534, -0.0099081601947546, -0.01734575815498829, -0.022899799048900604, -0.00996096059679985, -0.018402311950922012, -0.06909636408090591, -0.0034...
Constrained fractional set programs and their application in local clustering and community detection
https://proceedings.mlr.press/v28/buhler13.html
[ "Thomas Bühler", "Shyam Sundar Rangapuram", "Simon Setzer", "Matthias Hein" ]
null
null
The (constrained) minimization of a ratio of set functions is a problem frequently occurring in clustering and community detection. As these optimization problems are typically NP-hard, one uses convex or spectral relaxations in practice. While these relaxations can be solved globally optimally, they are often too loos...
[]
null
71
1306.3409
title_snapshot
[ -0.005977143999189138, -0.0247337706387043, -0.005045575555413961, 0.05500061810016632, 0.05442614480853081, 0.04321908578276634, 0.013132342137396336, -0.00007122513488866389, -0.027305537834763527, -0.04217829182744026, -0.0408618189394474, 0.01618739403784275, -0.07814915478229523, -0.0...
Efficient Semi-supervised and Active Learning of Disjunctions
https://proceedings.mlr.press/v28/balcan13.html
[ "Nina Balcan", "Christopher Berlind", "Steven Ehrlich", "Yingyu Liang" ]
null
null
We provide efficient algorithms for learning disjunctions in the semi-supervised setting under a natural regularity assumption introduced by (Balcan & Blum, 2005). We prove bounds on the sample complexity of our algorithms under a mild restriction on the data distribution. We also give an active learning algorithm with...
[]
null
72
null
null
[ 0.008724300190806389, -0.014482911676168442, -0.0325394906103611, 0.0440814383327961, 0.035386741161346436, -0.0071688429452478886, 0.013705817051231861, -0.010239537805318832, -0.0101081607863307, -0.009765667840838432, -0.0013468939578160644, 0.016569633036851883, -0.09350578486919403, 0...
Convex Adversarial Collective Classification
https://proceedings.mlr.press/v28/torkamani13.html
[ "MohamadAli Torkamani", "Daniel Lowd" ]
null
null
In this paper, we present a novel method for robustly performing collective classification in the presence of a malicious adversary that can modify up to a fixed number of binary-valued attributes. Our method is formulated as a convex quadratic program that guarantees optimal weights against a worst-case adversary...
[]
null
73
null
null
[ 0.008135736919939518, -0.02581547014415264, -0.007528862915933132, 0.055322639644145966, -0.0014803028898313642, 0.01815149374306202, 0.03571448102593422, -0.013554548844695091, -0.0269932858645916, -0.030024396255612373, -0.024518262594938278, 0.006267385557293892, -0.07648461312055588, -...
Rounding Methods for Discrete Linear Classification
https://proceedings.mlr.press/v28/chevaleyre13.html
[ "Yann Chevaleyre", "Frédéerick Koriche", "Jean-daniel Zucker" ]
null
null
Learning discrete linear functions is a notoriously difficult challenge. In this paper, the learning task is cast as combinatorial optimization problem: given a set of positive and negative feature vectors in the Euclidean space, the goal is to find a discrete linear function that minimizes the cumulative hinge loss of...
[]
null
74
null
null
[ -0.011484615504741669, 0.0028834855183959007, -0.012440062128007412, 0.02069382183253765, 0.036421481519937515, 0.06523742526769638, 0.0074344417080283165, -0.03862917050719261, -0.03945605456829071, -0.027497202157974243, -0.012317291460931301, -0.02250758558511734, -0.09550606459379196, ...
Mixture of Mutually Exciting Processes for Viral Diffusion
https://proceedings.mlr.press/v28/yang13a.html
[ "Shuang-Hong Yang", "Hongyuan Zha" ]
null
null
\emphDiffusion network inference and \emphmeme tracking have been two key challenges in viral diffusion. This paper shows that these two tasks can be addressed simultaneously with a probabilistic model involving a mixture of mutually exciting point processes. A fast learning algorithms is developed based on mean-field ...
[]
null
75
null
null
[ 0.026155443862080574, -0.0017536419909447432, -0.008748888038098812, 0.023570183664560318, 0.035257693380117416, 0.017423119395971298, 0.023242376744747162, 0.04068369045853615, -0.025664038956165314, -0.05149137228727341, 0.02405247464776039, -0.010429944843053818, -0.043725382536649704, ...
Gaussian Process Vine Copulas for Multivariate Dependence
https://proceedings.mlr.press/v28/lopez-paz13.html
[ "David Lopez-Paz", "Jose Miguel Hernández-Lobato", "Ghahramani Zoubin" ]
null
null
Copulas allow to learn marginal distributions separately from the multivariate dependence structure (copula) that links them together into a density function. Vine factorizations ease the learning of high-dimensional copulas by constructing a hierarchy of conditional bivariate copulas. However, to simplify inference, ...
[]
null
76
1302.3979
title_snapshot
[ 0.008438914082944393, 0.008705442771315575, -0.002301917178556323, 0.012727359309792519, 0.05924065038561821, 0.04517075791954994, -0.0012699236394837499, -0.0218364130705595, -0.005103105679154396, -0.049968235194683075, 0.01288839615881443, 0.03822208568453789, -0.06776423007249832, 0.03...
Stochastic Simultaneous Optimistic Optimization
https://proceedings.mlr.press/v28/valko13.html
[ "Michal Valko", "Alexandra Carpentier", "Rémi Munos" ]
null
null
We study the problem of global maximization of a function f given a finite number of evaluations perturbed by noise. We consider a very weak assumption on the function, namely that it is locally smooth (in some precise sense) with respect to some semi-metric, around one of its global maxima. Compared to previous works ...
[]
null
77
2604.24537
title_snapshot
[ -0.02037963457405567, 0.00956796482205391, 0.023059288039803505, 0.05166212096810341, 0.026424651965498924, 0.05661223456263542, 0.01888633705675602, 0.028656674548983574, -0.012118907645344734, -0.03807558864355087, 0.004963855724781752, 0.005015050992369652, -0.07503111660480499, -0.0395...
Toward Optimal Stratification for Stratified Monte-Carlo Integration
https://proceedings.mlr.press/v28/carpentier13.html
[ "Alexandra Carpentier", "Rémi Munos" ]
null
null
We consider the problem of adaptive stratified sampling for Monte Carlo integration of a function, given a finite number of function evaluations perturbed by noise. Here we address the problem of adapting simultaneously the number of samples into each stratum and the stratification itself. We show a tradeoff in the siz...
[]
null
78
1303.2892
title_snapshot
[ -0.041774798184633255, -0.004677476827055216, -0.008455615490674973, 0.05367639660835266, 0.047072187066078186, 0.03886137530207634, 0.004845743998885155, -0.03494381904602051, -0.026986537501215935, -0.04879121854901314, -0.0029841973446309566, 0.005541933234781027, -0.06044115126132965, ...
A General Iterative Shrinkage and Thresholding Algorithm for Non-convex Regularized Optimization Problems
https://proceedings.mlr.press/v28/gong13a.html
[ "Pinghua Gong", "Changshui Zhang", "Zhaosong Lu", "Jianhua Huang", "Jieping Ye" ]
null
null
Non-convex sparsity-inducing penalties have recently received considerable attentions in sparse learning. Recent theoretical investigations have demonstrated their superiority over the convex counterparts in several sparse learning settings. However, solving the non-convex optimization problems associated with non-conv...
[]
null
79
1303.4434
title_snapshot
[ -0.040457382798194885, -0.020357899367809296, 0.0069771562702953815, 0.03113478235900402, 0.030106140300631523, 0.04002010449767113, 0.03584682196378708, 0.002596938982605934, -0.05083208903670311, -0.0492699109017849, -0.0013278129044920206, -0.006638288032263517, -0.03680603578686714, -0...
Thurstonian Boltzmann Machines: Learning from Multiple Inequalities
https://proceedings.mlr.press/v28/tran13.html
[ "Truyen Tran", "Dinh Phung", "Svetha Venkatesh" ]
null
null
We introduce Thurstonian Boltzmann Machines (TBM), a unified architecture that can naturally incorporate a wide range of data inputs at the same time. Our motivation rests in the Thurstonian view that many discrete data types can be considered as being generated from a subset of underlying latent continuous variables, ...
[]
null
80
1408.0055
title_snapshot
[ 0.01188050676137209, -0.013192109763622284, -0.02912776917219162, 0.02533119171857834, 0.005850342568010092, 0.015471263788640499, 0.01881950907409191, 0.00335397245362401, -0.04061787575483322, -0.0157412551343441, 0.0042882817797362804, -0.00946186762303114, -0.06156989187002182, 0.00337...
A Variational Approximation for Topic Modeling of Hierarchical Corpora
https://proceedings.mlr.press/v28/kim13.html
[ "Do-kyum Kim", "Geoffrey Voelker", "Lawrence Saul" ]
null
null
We study the problem of topic modeling in corpora whose documents are organized in a multi-level hierarchy. We explore a parametric approach to this problem, assuming that the number of topics is known or can be estimated by cross-validation. The models we consider can be viewed as special (finite-dimensional) instan...
[]
null
81
null
null
[ -0.02340344712138176, 0.006947777234017849, -0.0063203019089996815, 0.05455470457673073, 0.037738919258117676, 0.016987444832921028, 0.012975270859897137, -0.007071472704410553, -0.021183857694268227, -0.014505751430988312, -0.023364830762147903, 0.009559868834912777, -0.0624588280916214, ...
Forecastable Component Analysis
https://proceedings.mlr.press/v28/goerg13.html
[ "Georg Goerg" ]
null
null
I introduce Forecastable Component Analysis (ForeCA), a novel dimension reduction technique for temporally dependent signals. Based on a new forecastability measure, ForeCA finds an optimal transformation to separate a multivariate time series into a forecastable and an orthogonal white noise space. I present a converg...
[]
null
82
null
null
[ -0.01996872015297413, -0.033600274473428726, 0.03450870141386986, 0.00045649585081264377, 0.05036832392215729, 0.05289365351200104, 0.02929486706852913, 0.015833789482712746, -0.03176987171173096, -0.05006078630685806, 0.01786288432776928, 0.0012334504863247275, -0.04058651626110077, 0.012...
Ellipsoidal Multiple Instance Learning
https://proceedings.mlr.press/v28/krummenacher13.html
[ "Gabriel Krummenacher", "Cheng Soon Ong", "Joachim Buhmann" ]
null
null
We propose a large margin method for asymmetric learning with ellipsoids, called eMIL, suited to multiple instance learning (MIL). We derive the distance between ellipsoids and the hyperplane, generalising the standard support vector machine. Negative bags in MIL contain only negative instances, and we treat them akin ...
[]
null
83
null
null
[ -0.029516324400901794, 0.0038986883591860533, 0.02443275786936283, 0.02989049255847931, 0.004262368194758892, 0.044380832463502884, 0.01576142944395542, -0.00949123501777649, -0.04358547925949097, -0.02463596500456333, -0.004522203467786312, 0.019393937662243843, -0.07450954616069794, -0.0...
Local Low-Rank Matrix Approximation
https://proceedings.mlr.press/v28/lee13.html
[ "Joonseok Lee", "Seungyeon Kim", "Guy Lebanon", "Yoram Singer" ]
null
null
Matrix approximation is a common tool in recommendation systems, text mining, and computer vision. A prevalent assumption in constructing matrix approximations is that the partially observed matrix is of low-rank. We propose a new matrix approximation model where we assume instead that the matrix is locally of low-rank...
[]
null
84
null
null
[ -0.004830601159483194, -0.023356685414910316, 0.023937180638313293, 0.019735414534807205, 0.041413187980651855, 0.016740819439291954, -0.00032496039057150483, -0.018574180081486702, -0.025073571130633354, -0.024845639243721962, -0.007884446531534195, 0.0060961488634347916, -0.068542897701263...
Generic Exploration and K-armed Voting Bandits
https://proceedings.mlr.press/v28/urvoy13.html
[ "Tanguy Urvoy", "Fabrice Clerot", "Raphael Féraud", "Sami Naamane" ]
null
null
We study a stochastic online learning scheme with partial feedback where the utility of decisions is only observable through an estimation of the environment parameters. We propose a generic pure-exploration algorithm, able to cope with various utility functions from multi-armed bandits settings to dueling bandits. The...
[]
null
85
null
null
[ -0.01684257760643959, -0.014308408834040165, 0.0025042493361979723, 0.038076259195804596, 0.004393648821860552, 0.03145970404148102, 0.032532207667827606, 0.00794121716171503, -0.03380349650979042, -0.014513724483549595, -0.021097252145409584, 0.010073539800941944, -0.06148263439536095, -0...
A unifying framework for vector-valued manifold regularization and multi-view learning
https://proceedings.mlr.press/v28/haquang13.html
[ "Minh Hà Quang", "Loris Bazzani", "Vittorio Murino" ]
null
null
This paper presents a general vector-valued reproducing kernel Hilbert spaces (RKHS) formulation for the problem of learning an unknown functional dependency between a structured input space and a structured output space, in the Semi-Supervised Learning setting. Our formulation includes as special cases Vector-valued ...
[]
null
86
1401.8066
title_judge
[ -0.012670603580772877, -0.03708159551024437, 0.02997390739619732, 0.026455793529748917, 0.029881807044148445, 0.06093522906303406, 0.011221683584153652, -0.02148306928575039, -0.019200952723622322, -0.05657079443335533, -0.026862220838665962, 0.02621549554169178, -0.08170103281736374, 0.01...
Learning Connections in Financial Time Series
https://proceedings.mlr.press/v28/ganeshapillai13.html
[ "Gartheeban Ganeshapillai", "John Guttag", "Andrew Lo" ]
null
null
To reduce risk, investors seek assets that have high expected return and are unlikely to move in tandem. Correlation measures are generally used to quantify the connections between equities. The 2008 financial crisis, and its aftermath, demonstrated the need for a better way to quantify these connections. We present a ...
[]
null
87
null
null
[ 0.005048504564911127, -0.022770337760448456, -0.0018450660863891244, 0.03748900815844536, 0.056066323071718216, 0.031467895954847336, 0.008200784213840961, 0.01019788347184658, -0.01422474067658186, -0.02731945924460888, 0.00282878614962101, 0.032393679022789, -0.06829799711704254, -0.0003...
Fast dropout training
https://proceedings.mlr.press/v28/wang13a.html
[ "Sida Wang", "Christopher Manning" ]
null
null
Preventing feature co-adaptation by encouraging independent contributions from different features often improves classification and regression performance. Dropout training (Hinton et al., 2012) does this by randomly dropping out (zeroing) hidden units and input features during training of neural networks. However, re...
[]
null
88
null
null
[ 0.0021706183906644583, -0.031058834865689278, -0.018905512988567352, 0.05887022987008095, 0.02776505798101425, 0.030230028554797173, 0.005774341523647308, -0.011284569278359413, -0.026837574318051338, -0.024393433704972267, -0.058563027530908585, 0.021760817617177963, -0.0674859955906868, ...
Scalable Optimization of Neighbor Embedding for Visualization
https://proceedings.mlr.press/v28/yang13b.html
[ "Zhirong Yang", "Jaakko Peltonen", "Samuel Kaski" ]
null
null
Neighbor embedding (NE) methods have found their use in data visualization but are limited in big data analysis tasks due to their O(n^2) complexity for n data samples. We demonstrate that the obvious approach of subsampling produces inferior results and propose a generic approximated optimization technique that reduce...
[]
null
89
null
null
[ -0.016689442098140717, -0.026833001524209976, 0.028422944247722626, 0.006872239988297224, 0.0382574126124382, 0.04684193804860115, 0.0020083903800696135, -0.03347774222493172, -0.03393949568271637, -0.054210811853408813, -0.005242368672043085, -0.025166591629385948, -0.06316707283258438, 0...
Precision-recall space to correct external indices for biclustering
https://proceedings.mlr.press/v28/hanczar13.html
[ "Blaise Hanczar", "Mohamed Nadif" ]
null
null
Biclustering is a major tool of data mining in many domains and many algorithms have emerged in recent years. All these algorithms aim to obtain coherent biclusters and it is crucial to have a reliable procedure for their validation. We point out the problem of size bias in biclustering evaluation and show how it can l...
[]
null
90
null
null
[ -0.03187190741300583, 0.00426197936758399, -0.02394460141658783, 0.008253710344433784, 0.06956306099891663, 0.012454641051590443, 0.003117342945188284, 0.007107910700142384, -0.03658978268504143, -0.04886813089251518, 0.01759255677461624, -0.01878603734076023, -0.05628956854343414, 0.00431...
Monochromatic Bi-Clustering
https://proceedings.mlr.press/v28/wulff13.html
[ "Sharon Wulff", "Ruth Urner", "Shai Ben-David" ]
null
null
We propose a natural cost function for the bi-clustering task, the monochromatic cost. This cost function is suitable for detecting meaningful homogeneous bi-clusters based on categorical valued input matrices. Such tasks arise in many applications, such as the analysis of social networks and in systems-biology where ...
[]
null
91
null
null
[ -0.009434178471565247, -0.003352490020915866, -0.02081017941236496, 0.044736556708812714, 0.03931434452533722, 0.042213041335344315, 0.006684407126158476, 0.0050191087648272514, -0.020614301785826683, -0.040793146938085556, -0.00003680068766698241, -0.010601772926747799, -0.07447361201047897...
Gated Autoencoders with Tied Input Weights
https://proceedings.mlr.press/v28/alain13.html
[ "Droniou Alain", "Sigaud Olivier" ]
null
null
The semantic interpretation of images is one of the core applications of deep learning. Several techniques have been recently proposed to model the relation between two images, with application to pose estimation, action recognition or invariant object recognition. Among these techniques, higher-order Boltzmann machine...
[]
null
92
null
null
[ 0.016061535105109215, -0.01625969260931015, -0.008084584027528763, 0.04472576826810837, 0.023054728284478188, 0.03180346637964249, 0.016814015805721283, -0.029160059988498688, -0.00782655831426382, -0.039281655102968216, -0.014397673308849335, -0.00556239765137434, -0.07760696113109589, -0...
Strict Monotonicity of Sum of Squares Error and Normalized Cut in the Lattice of Clusterings
https://proceedings.mlr.press/v28/rebagliati13.html
[ "Nicola Rebagliati" ]
null
null
Sum of Squares Error and Normalized Cut are two widely used clustering functional. It is known their minimum values are monotone with respect to the input number of clusters and this monotonicity does not allow for a simple automatic selection of a correct number of clusters. Here we study monotonicity not just on the ...
[]
null
93
null
null
[ -0.023537470027804375, 0.0021181663032621145, 0.005196890328079462, 0.03597581759095192, 0.042378704994916916, 0.045689959079027176, 0.027920763939619064, -0.025399429723620415, -0.027835099026560783, -0.041319459676742554, -0.027031322941184044, -0.015779152512550354, -0.05290818586945534, ...
Transition Matrix Estimation in High Dimensional Time Series
https://proceedings.mlr.press/v28/han13a.html
[ "Fang Han", "Han Liu" ]
null
null
In this paper, we propose a new method in estimating transition matrices of high dimensional vector autoregressive (VAR) models. Here the data are assumed to come from a stationary Gaussian VAR time series. By formulating the problem as a linear program, we provide a new approach to conduct inference on such models. In...
[]
null
94
null
null
[ -0.02161613665521145, -0.03788226470351219, 0.0047213430516421795, -0.01710064522922039, 0.012578826397657394, 0.015764031559228897, 0.0574323944747448, 0.024989958852529526, -0.014083468355238438, -0.037221815437078476, 0.01451394334435463, -0.0024082332383841276, -0.03710299730300903, 0....
Label Partitioning For Sublinear Ranking
https://proceedings.mlr.press/v28/weston13.html
[ "Jason Weston", "Ameesh Makadia", "Hector Yee" ]
null
null
We consider the case of ranking a very large set of labels, items, or documents, which is common to information retrieval, recommendation, and large-scale annotation tasks. We present a general approach for converting an algorithm which has linear time in the size of the set to a sublinear one via label partitioning. O...
[]
null
95
null
null
[ -0.028940143063664436, -0.04492057114839554, -0.0031843555625528097, 0.022432418540120125, 0.0313580147922039, 0.0045675053261220455, -0.0018545291386544704, -0.019100630655884743, -0.008693963289260864, -0.009928958490490913, -0.02066430076956749, 0.011278180405497551, -0.06303565204143524,...
Subproblem-Tree Calibration: A Unified Approach to Max-Product Message Passing
https://proceedings.mlr.press/v28/wang13b.html
[ "Huayan Wang", "Koller Daphne" ]
null
null
Max-product (max-sum) message passing algorithms are widely used for MAP inference in MRFs. It has many variants sharing a common flavor of passing "messages" over some graph-object. Recent advances revealed that its convergent versions (such as MPLP, MSD, TRW-S) can be viewed as performing block coordinate descent (BC...
[]
null
96
null
null
[ -0.026705622673034668, -0.004736465401947498, -0.0029389457777142525, 0.0349055640399456, 0.03786780312657356, 0.02777113765478134, 0.02382141910493374, 0.015646735206246376, -0.009729764424264431, -0.04563404247164726, 0.012169701978564262, -0.01635221764445305, -0.0741945281624794, 0.009...
Collaborative hyperparameter tuning
https://proceedings.mlr.press/v28/bardenet13.html
[ "Rémi Bardenet", "Mátyás Brendel", "Balázs Kégl", "Michèle Sebag" ]
null
null
Hyperparameter learning has traditionally been a manual task because of the limited number of trials. Today’s computing infrastructures allow bigger evaluation budgets, thus opening the way for algorithmic approaches. Recently, surrogate-based optimization was successfully applied to hyperparameter learning for deep be...
[]
null
97
null
null
[ -0.000075713345722761, -0.03832869604229927, -0.013189062476158142, 0.039326686412096024, 0.014194684103131294, 0.03834269940853119, 0.06853342801332474, -0.03467213362455368, 0.01798289828002453, -0.05317485332489014, 0.009100525639951229, 0.0009260636288672686, -0.04275527223944664, -0.0...
SADA: A General Framework to Support Robust Causation Discovery
https://proceedings.mlr.press/v28/cai13.html
[ "Ruichu Cai", "Zhenjie Zhang", "Zhifeng Hao" ]
null
null
Causality discovery without manipulation is considered a crucial problem to a variety of applications, such as genetic therapy. The state-of-the-art solutions, e.g. LiNGAM, return accurate results when the number of labeled samples is larger than the number of variables. These approaches are thus applicable only when l...
[]
null
98
null
null
[ 0.004369412083178759, -0.01984107866883278, -0.037902865558862686, 0.029949529096484184, 0.044759418815374374, 0.04292618855834007, 0.04424115642905235, -0.0032963119447231293, -0.017624402418732643, -0.04204750061035156, 0.029093973338603973, 0.0019770690705627203, -0.06206655874848366, 0...
Learning and Selecting Features Jointly with Point-wise Gated Boltzmann Machines
https://proceedings.mlr.press/v28/sohn13.html
[ "Kihyuk Sohn", "Guanyu Zhou", "Chansoo Lee", "Honglak Lee" ]
null
null
Unsupervised feature learning has emerged as a promising tool in learning representations from unlabeled data. However, it is still challenging to learn useful high-level features when the data contains a significant amount of irrelevant patterns. Although feature selection can be used for such complex data, it may fai...
[]
null
99
null
null
[ 0.000030027007596800104, -0.019219936802983284, 0.0020189438946545124, 0.01925213634967804, 0.026447227224707603, 0.015074191614985466, 0.01694810576736927, -0.0256227795034647, -0.0025112167932093143, -0.032258741557598114, -0.013738890178501606, -0.008845881558954716, -0.06332361698150635,...
Sequential Bayesian Search
https://proceedings.mlr.press/v28/wen13.html
[ "Zheng Wen", "Branislav Kveton", "Brian Eriksson", "Sandilya Bhamidipati" ]
null
null
Millions of people search daily for movies, music, and books on the Internet. Unfortunately, non-personalized exploration of items can result in an infeasible number of costly interaction steps. We study the problem of efficient, repeated interactive search. In this problem, the user is navigated to the items of intere...
[]
null
100
null
null
[ -0.017155364155769348, 0.003914672415703535, -0.005205248016864061, 0.04589073359966278, 0.05365234985947609, -0.01549129281193018, 0.01949567161500454, 0.014148915186524391, -0.0066305166110396385, -0.029999185353517532, -0.03500989452004433, 0.037841252982616425, -0.04395272955298424, -0...