Proceedings of the 31st International Conference on Machine Learning, Cycle 1
ICML c1, 2014.

@proceedings{ICML-c1-2014,
	address       = "Beijing, China",
	ee            = "http://jmlr.org/proceedings/papers/v32/",
	publisher     = "{JMLR.org}",
	series        = "{JMLR Proceedings}",
	title         = "{Proceedings of the 31st International Conference on Machine Learning, Cycle 1}",
	volume        = 32,
	year          = 2014,
}

Contents (85 items)

14 ×#learning
10 ×#multi
9 ×#performance
8 ×#algorithm
8 ×#modelling
7 ×#kernel
7 ×#online
6 ×#estimation
5 ×#bound
5 ×#clustering

ICML-c1-2014-SamdaniCR #clustering #online

A Discriminative Latent Variable Model for Online Clustering (RS, KWC, DR), pp. 1–9.

ICML-c1-2014-MuandetFSGS #estimation #kernel

Kernel Mean Estimation and Stein Effect (KM, KF, BKS, AG, BS), pp. 10–18.

ICML-c1-2014-SteegGSD #clustering

Demystifying Information-Theoretic Clustering (GVS, AG, FS, SD), pp. 19–27.

ICML-c1-2014-ZhangHL #heuristic #performance

Covering Number for Efficient Heuristic-based POMDP Planning (ZZ, DH, WSL), pp. 28–36.

ICML-c1-2014-YangSX #classification

The Coherent Loss Function for Classification (WY, MS, HX), pp. 37–45.

ICML-c1-2014-ZhongK #multi #performance #probability

Fast Stochastic Alternating Direction Method of Multipliers (WZ, JTYK), pp. 46–54.

ICML-c1-2014-ChenSMKWK #adaptation #detection

Active Detection via Adaptive Submodularity (YC, HS, CFM, LPK, SW, AK), pp. 55–63.

ICML-c1-2014-Shalev-Shwartz0 #coordination #probability

Accelerated Proximal Stochastic Dual Coordinate Ascent for Regularized Loss Minimization (SSS, TZ), pp. 64–72.

ICML-c1-2014-LinX #adaptation #continuation #optimisation

An Adaptive Accelerated Proximal Gradient Method and its Homotopy Continuation for Sparse Optimization (QL, LX), pp. 73–81.

ICML-c1-2014-PinheiroC #network

Recurrent Convolutional Neural Networks for Scene Labeling (PHOP, RC), pp. 82–90.

ICML-c1-2014-MaMY #algorithm #statistics

A Statistical Perspective on Algorithmic Leveraging (PM, MWM, BY), pp. 91–99.

ICML-c1-2014-GopalanMM #online #problem

Thompson Sampling for Complex Online Problems (AG, SM, YM), pp. 100–108.

ICML-c1-2014-TaiebH #multi

Boosting multi-step autoregressive forecasts (SBT, RJH), pp. 109–117.

ICML-c1-2014-RajkumarA #algorithm #convergence #rank #statistics

A Statistical Convergence Perspective of Algorithms for Rank Aggregation from Pairwise Data (AR, SA), pp. 118–126.

ICML-c1-2014-MannM #approximate #policy #scalability

Scaling Up Approximate Value Iteration with Options: Better Policies with Fewer Iterations (TAM, SM), pp. 127–135.

ICML-c1-2014-MaillardM

Latent Bandits (OAM, SM), pp. 136–144.

ICML-c1-2014-NguyenB #performance #process

Fast Allocation of Gaussian Process Experts (TVN, EVB), pp. 145–153.

ICML-c1-2014-GopalY #clustering #modelling

Von Mises-Fisher Clustering Models (SG, YY), pp. 154–162.

ICML-c1-2014-ChazalGLM #convergence #data analysis #diagrams #estimation #persistent

Convergence rates for persistence diagram estimation in Topological Data Analysis (FC, MG, CL, BM), pp. 163–171.

ICML-c1-2014-GiesekeHOI #nearest neighbour #query

Buffer k-d Trees: Processing Massive Nearest Neighbor Queries on GPUs (FG, JH, CEO, CI), pp. 172–180.

ICML-c1-2014-BalanCW

Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget (AKB, YC, MW), pp. 181–189.

ICML-c1-2014-TangMNMZ #analysis #comprehension #modelling #topic

Understanding the Limiting Factors of Topic Modeling via Posterior Contraction Analysis (JT, ZM, XN, QM, MZ), pp. 190–198.

ICML-c1-2014-RabinovichB #topic

The Inverse Regression Topic Model (MR, DMB), pp. 199–207.

ICML-c1-2014-ChanA #consistency #graph #modelling

A Consistent Histogram Estimator for Exchangeable Graph Models (SHC, EA), pp. 208–216.

ICML-c1-2014-LethamSS #transaction

Latent Variable Copula Inference for Bundle Pricing from Retail Transaction Data (BL, WS, AS), pp. 217–225.

ICML-c1-2014-LuoS #learning #online #towards

Towards Minimax Online Learning with Unknown Time Horizon (HL, RES), pp. 226–234.

ICML-c1-2014-MillerBAG #analysis #process

Factorized Point Process Intensities: A Spatial Analysis of Professional Basketball (AM, LB, RPA, KG), pp. 235–243.

ICML-c1-2014-RamdasP #kernel

Margins, Kernels and Non-linear Smoothed Perceptrons (AR, JP), pp. 244–252.

ICML-c1-2014-MeiZZ #first-order #logic #modelling #robust

Robust RegBayes: Selectively Incorporating First-Order Logic Domain Knowledge into Bayesian Models (SM, JZ, JZ), pp. 253–261.

ICML-c1-2014-MohriM #algorithm #learning #optimisation

Learning Theory and Algorithms for revenue optimization in second price auctions with reserve (MM, AMM), pp. 262–270.

ICML-c1-2014-ErmonGSS #constraints #integration

Low-density Parity Constraints for Hashing-Based Discrete Integration (SE, CPG, AS, BS), pp. 271–279.

ICML-c1-2014-SeldinBCA #multi #predict

Prediction with Limited Advice and Multiarmed Bandits with Paid Observations (YS, PLB, KC, YAY), pp. 280–287.

ICML-c1-2014-NguyenPNVB #clustering #multi #parametricity

Bayesian Nonparametric Multilevel Clustering with Group-Level Contexts (TVN, DQP, XN, SV, HB), pp. 288–296.

ICML-c1-2014-LajugieBA #clustering #learning #metric #problem

Large-Margin Metric Learning for Constrained Partitioning Problems (RL, FRB, SA), pp. 297–305.

ICML-c1-2014-SolomonRGB #learning

Wasserstein Propagation for Semi-Supervised Learning (JS, RMR, LJG, AB), pp. 306–314.

ICML-c1-2014-ZhangZZ #infinity #markov #modelling

Max-Margin Infinite Hidden Markov Models (AZ, JZ, BZ), pp. 315–323.

ICML-c1-2014-LiuJL #approximate #kernel #performance #using

Efficient Approximation of Cross-Validation for Kernel Methods using Bouligand Influence Function (YL, SJ, SL), pp. 324–332.

ICML-c1-2014-SinghP #difference #estimation #exponential

Generalized Exponential Concentration Inequality for Renyi Divergence Estimation (SS, BP), pp. 333–341.

ICML-c1-2014-ChenLL #multi #online #problem

Boosting with Online Binary Learners for the Multiclass Bandit Problem (STC, HTL, CJL), pp. 342–350.

ICML-c1-2014-SomaKIK #algorithm #performance

Optimal Budget Allocation: Theoretical Guarantee and Efficient Algorithm (TS, NK, KI, KiK), pp. 351–359.

ICML-c1-2014-SoufianiPX #modelling #parametricity #ranking

Computing Parametric Ranking Models via Rank-Breaking (HAS, DCP, LX), pp. 360–368.

ICML-c1-2014-Abbasi-YadkoriBK

Tracking Adversarial Targets (YAY, PLB, VK), pp. 369–377.

ICML-c1-2014-ShiZ #learning #online

Online Bayesian Passive-Aggressive Learning (TS, JZ), pp. 378–386.

ICML-c1-2014-SilverLHDWR #algorithm #policy

Deterministic Policy Gradient Algorithms (DS, GL, NH, TD, DW, MAR), pp. 387–395.

ICML-c1-2014-LianREC #correlation #markov #modelling #process

Modeling Correlated Arrival Events with Latent Semi-Markov Processes (WL, VR, BE, LC), pp. 396–404.

ICML-c1-2014-BardenetDH #adaptation #approach #markov #monte carlo #scalability #towards

Towards scaling up Markov chain Monte Carlo: an adaptive subsampling approach (RB, AD, CCH), pp. 405–413.

ICML-c1-2014-CicaleseLS #optimisation #testing

Diagnosis determination: decision trees optimizing simultaneously worst and expected testing cost (FC, ESL, AMS), pp. 414–422.

ICML-c1-2014-LiL #classification #multi

Condensed Filter Tree for Cost-Sensitive Multi-Label Classification (CLL, HTL), pp. 423–431.

ICML-c1-2014-OrabonaHSJ #on the #random

On Measure Concentration of Random Maximum A-Posteriori Perturbations (FO, TH, ADS, TSJ), pp. 432–440.

ICML-c1-2014-Thomas #algorithm #bias

Bias in Natural Actor-Critic Algorithms (PT), pp. 441–448.

ICML-c1-2014-DenisGH #bound #learning #matrix

Dimension-free Concentration Bounds on Hankel Matrices for Spectral Learning (FD, MG, AH), pp. 449–457.

ICML-c1-2014-LiWLZT #dependence #modelling #on the #topic

On Modelling Non-linear Topical Dependencies (ZL, SW, JL, PZ, JT), pp. 458–466.

ICML-c1-2014-UriaML

A Deep and Tractable Density Estimator (BU, IM, HL), pp. 467–475.

ICML-c1-2014-JainT #bound #independence #learning

(Near) Dimension Independent Risk Bounds for Differentially Private Learning (PJ, AGT), pp. 476–484.

ICML-c1-2014-YangSAM #invariant #kernel #monte carlo

Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels (JY, VS, HA, MWM), pp. 485–493.

ICML-c1-2014-KarampatziakisM

Discriminative Features via Generalized Eigenvectors (NK, PM), pp. 494–502.

ICML-c1-2014-LiuYF #algorithm #constraints

Forward-Backward Greedy Algorithms for General Convex Smooth Functions over A Cardinality Constraint (JL, JY, RF), pp. 503–511.

ICML-c1-2014-DickGS #learning #markov #online #process #sequence

Online Learning in Markov Decision Processes with Changing Cost Sequences (TD, AG, CS), pp. 512–520.

ICML-c1-2014-CombesP #algorithm #bound

Unimodal Bandits: Regret Lower Bounds and Optimal Algorithms (RC, AP), pp. 521–529.

ICML-c1-2014-IyerNS #bound #convergence #estimation #kernel

Maximum Mean Discrepancy for Class Ratio Estimation: Convergence Bounds and Kernel Selection (AI, SN, SS), pp. 530–538.

ICML-c1-2014-KhaleghiR #consistency #estimation

Asymptotically consistent estimation of the number of change points in highly dependent time series (AK, DR), pp. 539–547.

ICML-c1-2014-ShalitC #coordination #learning #matrix #orthogonal

Coordinate-descent for learning orthogonal matrices through Givens rotations (US, GC), pp. 548–556.

ICML-c1-2014-Shrivastava0 #performance #permutation

Densifying One Permutation Hashing via Rotation for Fast Near Neighbor Search (AS, PL), pp. 557–565.

ICML-c1-2014-HsiehSD #divide and conquer #kernel

A Divide-and-Conquer Solver for Kernel Support Vector Machines (CJH, SS, ISD), pp. 566–574.

ICML-c1-2014-HsiehO

Nuclear Norm Minimization via Active Subspace Selection (CJH, PAO), pp. 575–583.

ICML-c1-2014-AroraBGM #bound #learning

Provable Bounds for Learning Some Deep Representations (SA, AB, RG, TM), pp. 584–592.

ICML-c1-2014-Yu0KD #learning #multi #scalability

Large-scale Multi-label Learning with Missing Labels (HFY, PJ, PK, ISD), pp. 593–601.

ICML-c1-2014-TandonR #graph #learning

Learning Graphs with a Few Hubs (RT, PDR), pp. 602–610.

ICML-c1-2014-LacosteMLL #learning

Agnostic Bayesian Learning of Ensembles (AL, MM, FL, HL), pp. 611–619.

ICML-c1-2014-AzadiS #multi #probability #towards

Towards an optimal stochastic alternating direction method of multipliers (SA, SS), pp. 620–628.

ICML-c1-2014-LanZS #monte carlo

Spherical Hamiltonian Monte Carlo for Constrained Target Distributions (SL, BZ, BS), pp. 629–637.

ICML-c1-2014-HajiaghayiKWB #estimation #markov #performance

Efficient Continuous-Time Markov Chain Estimation (MH, BK, LW, ABC), pp. 638–646.

ICML-c1-2014-DonahueJVHZTD #named #recognition #visual notation

DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition (JD, YJ, OV, JH, NZ, ET, TD), pp. 647–655.

ICML-c1-2014-YogatamaS #multi #word

Making the Most of Bag of Words: Sentence Regularization with Alternating Direction Method of Multipliers (DY, NAS), pp. 656–664.

ICML-c1-2014-DenilMF #random

Narrowing the Gap: Random Forests In Theory and In Practice (MD, DM, NdF), pp. 665–673.

ICML-c1-2014-ChenBSW #matrix

Coherent Matrix Completion (YC, SB, SS, RW), pp. 674–682.

ICML-c1-2014-InouyeRD #dependence #topic #word

Admixture of Poisson MRFs: A Topic Model with Word Dependencies (DI, PDR, ISD), pp. 683–691.

ICML-c1-2014-SeijenS #online

True Online TD(λ) (HvS, RSS), pp. 692–700.

ICML-c1-2014-SiHD #approximate #kernel #memory management #performance

Memory Efficient Kernel Approximation (SS, CJH, ISD), pp. 701–709.

ICML-c1-2014-RooshenasL #interactive #learning #network

Learning Sum-Product Networks with Direct and Indirect Variable Interactions (AR, DL), pp. 710–718.

ICML-c1-2014-Sohl-DicksteinMD #monte carlo

Hamiltonian Monte Carlo Without Detailed Balance (JSD, MM, MRD), pp. 719–726.

ICML-c1-2014-SteinhardtL

Filtering with Abstract Particles (JS, PL), pp. 727–735.

ICML-c1-2014-Suzuki #coordination #multi #probability

Stochastic Dual Coordinate Ascent with Alternating Direction Method of Multipliers (TS), pp. 736–744.

ICML-c1-2014-ZhouT #generative #network #predict #probability

Deep Supervised and Convolutional Generative Stochastic Network for Protein Secondary Structure Prediction (JZ, OGT), pp. 745–753.

ICML-c1-2014-HutterHL #approach #performance

An Efficient Approach for Assessing Hyperparameter Importance (FH, HH, KLB), pp. 754–762.