Maria-Florina Balcan, Kilian Q. Weinberger
Proceedings of the 33rd International Conference on Machine Learning
ICML, 2016.

@proceedings{ICML-2016,
	editor        = "Maria-Florina Balcan and Kilian Q. Weinberger",
	ee            = "http://proceedings.mlr.press/v48/",
	publisher     = "{JMLR.org}",
	series        = "{JMLR Workshop and Conference Proceedings}",
	title         = "{Proceedings of the 33rd International Conference on Machine Learning}",
	volume        = 48,
	year          = 2016,
}

Contents (322 items)

ICML-2016-ShahZ #crowdsourcing #self

No Oops, You Won't Do It Again: Mechanisms for Self-correction in Crowdsourcing (NBS, DZ), pp. 1–10.

ICML-2016-ShahBGW #modelling #statistics #transitive

Stochastically Transitive Models for Pairwise Comparisons: Statistical and Computational Issues (NBS, SB, AG, MJW), pp. 11–20.

ICML-2016-Weller #modelling #visual notation

Uprooting and Rerooting Graphical Models (AW), pp. 21–29.

ICML-2016-ShahamCDJNCK #approach #learning

A Deep Learning Approach to Unsupervised Ensemble Learning (US, XC, OD, AJ, BN, JTC, YK), pp. 30–39.

ICML-2016-YangCS #graph #learning

Revisiting Semi-Supervised Learning with Graph Embeddings (ZY, WWC, RS), pp. 40–48.

ICML-2016-FinnLA #learning #optimisation #policy

Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization (CF, SL, PA), pp. 49–58.

ICML-2016-XieZX #learning #modelling

Diversity-Promoting Bayesian Learning of Latent Variable Models (PX, JZ0, EPX), pp. 59–68.

ICML-2016-KandasamyY #approximate #parametricity

Additive Approximations in High Dimensional Nonparametric Regression via the SALSA (KK, YY), pp. 69–78.

ICML-2016-LeeLO #probability #process

Hawkes Processes with Stochastic Excitations (YL, KWL, CSO), pp. 79–88.

ICML-2016-KhetanO #data-driven #performance #rank

Data-driven Rank Breaking for Efficient Rank Aggregation (AK, SO), pp. 89–98.

ICML-2016-BuloPK

Dropout distillation (SRB, LP, PK), pp. 99–107.

ICML-2016-FantiKORV

Metadata-conscious anonymous messaging (GCF, PK, SO, KR, PV), pp. 108–116.

ICML-2016-LiuZO #education #linear

The Teaching Dimension of Linear Learners (JL, XZ0, HO), pp. 117–126.

ICML-2016-CaragiannisPS

Truthful Univariate Estimators (IC, ADP, NS0), pp. 127–135.

ICML-2016-ArpitZNG #question #representation #why

Why Regularized Auto-Encoders learn Sparse Representation? (DA, YZ, HQN0, VG), pp. 136–144.

ICML-2016-NockCBN

k-variates++: more pluses in the k-means++ (RN, RC, RB, FN), pp. 145–154.

ICML-2016-RosenskiSS #approach #multi

Multi-Player Bandits - a Musical Chairs Approach (JR, OS, LS), pp. 155–163.

ICML-2016-SteegG

The Information Sieve (GVS, AG), pp. 164–172.

ICML-2016-AmodeiABCCCCCCD #recognition #speech

Deep Speech 2 : End-to-End Speech Recognition in English and Mandarin (DA, SA, RA, JB, EB, CC, JC, BC, JC, MC, AC, GD, EE, JHE, LF, CF, AYH, BJ, TH, PL, XL, LL, SN, AYN, SO, RP, SQ, JR, SS, DS, SS, CW0, YW, ZW, BX, YX, DY, JZ, ZZ), pp. 173–182.

ICML-2016-ZhangGR #consistency #feature model #on the

On the Consistency of Feature Selection With Lasso for Non-linear Targets (YZ, WG, SR), pp. 183–191.

ICML-2016-Metzen #multi #optimisation

Minimum Regret Search for Single- and Multi-Task Optimization (JHM), pp. 192–200.

ICML-2016-Gilad-BachrachD #named #network #throughput

CryptoNets: Applying Neural Networks to Encrypted Data with High Throughput and Accuracy (RGB, ND, KL, KEL, MN, JW), pp. 201–210.

ICML-2016-VladymyrovC #problem #scalability

The Variational Nystrom method for large-scale spectral problems (MV, MÁCP), pp. 211–220.

ICML-2016-LiOW #multi #network

Multi-Bias Non-linear Activation in Deep Neural Networks (HL, WO, XW0), pp. 221–229.

ICML-2016-LeeYH #learning #multi #symmetry

Asymmetric Multi-task Learning based on Task Relatedness and Confidence (GL, EY, SJH), pp. 230–238.

ICML-2016-Fan #estimation #fault #performance #robust

Accurate Robust and Efficient Error Estimation for Decision Trees (LF), pp. 239–247.

ICML-2016-Shamir #algorithm #convergence #performance #probability

Fast Stochastic Algorithms for SVD and PCA: Convergence Properties and Convexity (OS), pp. 248–256.

ICML-2016-Shamir16a #convergence #probability

Convergence of Stochastic Gradient Descent for PCA (OS), pp. 257–265.

ICML-2016-LanGBS #education #named

Dealbreaker: A Nonlinear Latent Variable Model for Educational Data (ASL, TG, RGB, CS), pp. 266–275.

ICML-2016-LiuLJ #kernel #testing

A Kernelized Stein Discrepancy for Goodness-of-fit Tests (QL, JDL, MIJ), pp. 276–284.

ICML-2016-XueEBGS #fourier

Variable Elimination in the Fourier Domain (YX, SE, RLB, CPG, BS), pp. 285–294.

ICML-2016-LiCLYSC #approximate #matrix #rank

Low-Rank Matrix Approximation with Stability (DL, CC0, QL, JY, LS, SMC), pp. 295–303.

ICML-2016-MenonO #estimation

Linking losses for density ratio and class-probability estimation (AKM, CSO), pp. 304–313.

ICML-2016-ReddiHSPS #optimisation #probability #reduction

Stochastic Variance Reduction for Nonconvex Optimization (SJR, AH, SS, BP, AJS), pp. 314–323.

ICML-2016-RanganathTB #modelling

Hierarchical Variational Models (RR, DT, DMB), pp. 324–333.

ICML-2016-AdamsSTPKM #data type #random #smarttech

Hierarchical Span-Based Conditional Random Fields for Labeling and Segmenting Events in Wearable Sensor Data Streams (RJA, NS, ET, AP, SK0, BMM), pp. 334–343.

ICML-2016-ChoromanskaCBJK

Binary embeddings with structured hashed projections (AC, KC, MB, TJ, SK, YL), pp. 344–353.

ICML-2016-MandtHB #algorithm #analysis #probability

A Variational Analysis of Stochastic Gradient Algorithms (SM, MDH, DMB), pp. 354–363.

ICML-2016-Gopal #adaptation

Adaptive Sampling for SGD by Exploiting Side Information (SG), pp. 364–372.

ICML-2016-YuL #learning #multi #performance

Learning from Multiway Data: Simple and Efficient Tensor Regression (RY, YL0), pp. 373–381.

ICML-2016-HoangHL #distributed #framework #modelling #parallel #process

A Distributed Variational Inference Framework for Unifying Parallel Sparse Gaussian Process Regression Models (TNH, QMH, BKHL), pp. 382–391.

ICML-2016-ZhangYJXZ #feedback #linear #online #optimisation #probability

Online Stochastic Linear Optimization under One-bit Feedback (LZ0, TY, RJ, YX, ZHZ), pp. 392–401.

ICML-2016-JenattonHA #adaptation #algorithm #constraints #online #optimisation

Adaptive Algorithms for Online Convex Optimization with Long-term Constraints (RJ, JCH, CA), pp. 402–411.

ICML-2016-SinglaTK #elicitation #learning

Actively Learning Hemimetrics with Applications to Eliciting User Preferences (AS, ST, AK0), pp. 412–420.

ICML-2016-ZarembaMJF #algorithm #learning

Learning Simple Algorithms from Examples (WZ, TM, AJ, RF), pp. 421–429.

ICML-2016-LererGF #learning #physics

Learning Physical Intuition of Block Towers by Example (AL, SG, RF), pp. 430–438.

ICML-2016-LiuSSF #learning #markov #network

Structure Learning of Partitioned Markov Networks (SL0, TS, MS, KF), pp. 439–448.

ICML-2016-YangZJY #learning #online

Tracking Slowly Moving Clairvoyant: Optimal Dynamic Regret of Online Learning with True and Noisy Gradient (TY, LZ0, RJ, JY), pp. 449–457.

ICML-2016-PodosinnikovaBL #modelling #multi

Beyond CCA: Moment Matching for Multi-View Models (AP, FRB, SLJ), pp. 458–467.

ICML-2016-UbaruS #matrix #performance #rank #scalability

Fast methods for estimating the Numerical rank of large matrices (SU, YS), pp. 468–477.

ICML-2016-XieGF #analysis #clustering

Unsupervised Deep Embedding for Clustering Analysis (JX, RBG, AF), pp. 478–487.

ICML-2016-Kasiviswanathan #empirical #learning #performance

Efficient Private Empirical Risk Minimization for High-dimensional Learning (SPK, HJ), pp. 488–497.

ICML-2016-VojnovicY #estimation #modelling #parametricity

Parameter Estimation for Generalized Thurstone Choice Models (MV, SYY), pp. 498–506.

ICML-2016-LiuWYY #network

Large-Margin Softmax Loss for Convolutional Neural Networks (WL, YW, ZY, MY0), pp. 507–516.

ICML-2016-CouilletWAS #approach #matrix #network #random

A Random Matrix Approach to Echo-State Neural Networks (RC, GW, HTA, HS), pp. 517–525.

ICML-2016-JohnsonZ #categorisation #using

Supervised and Semi-Supervised Text Categorization using LSTM for Region Embeddings (RJ, TZ0), pp. 526–534.

ICML-2016-OkOSY #classification #crowdsourcing

Optimality of Belief Propagation for Crowdsourced Classification (JO, SO, JS, YY), pp. 535–544.

ICML-2016-VinogradskaBNRS #modelling #process

Stability of Controllers for Gaussian Process Forward Models (JV, BB, DNT, AR, HS, JP0), pp. 545–554.

ICML-2016-HammCB #learning #multi

Learning privately from multiparty data (JH, YC, MB), pp. 555–563.

ICML-2016-WeiWRC #morphism #network

Network Morphism (TW, CW, YR, CWC), pp. 564–572.

ICML-2016-GrosseM #approximate #matrix

A Kronecker-factored approximate Fisher matrix for convolution layers (RBG, JM), pp. 573–582.

ICML-2016-RaviIJS #design #linear #modelling

Experimental Design on a Budget for Sparse Linear Models and Applications (SNR, VKI, SCJ, VS), pp. 583–592.

ICML-2016-OsokinALDL #optimisation

Minding the Gaps for Block Frank-Wolfe Optimization of Structured SVMs (AO, JBA, IL, PKD, SLJ), pp. 593–602.

ICML-2016-GaoLZ #crowdsourcing

Exact Exponent in Optimal Rates for Crowdsourcing (CG, YL, DZ), pp. 603–611.

ICML-2016-ZhangLL #classification #image #network #scalability

Augmenting Supervised Neural Networks with Unsupervised Objectives for Large-scale Image Classification (YZ, KL, HL), pp. 612–621.

ICML-2016-ShenLX #clustering #online #rank #taxonomy

Online Low-Rank Subspace Clustering by Basis Dictionary Pursuit (JS0, PL0, HX), pp. 622–631.

ICML-2016-Curtis #algorithm #optimisation #probability #self

A Self-Correcting Variable-Metric Algorithm for Stochastic Optimization (FC), pp. 632–641.

ICML-2016-SimsekliBCR #monte carlo #probability

Stochastic Quasi-Newton Langevin Monte Carlo (US, RB, ATC, GR), pp. 642–651.

ICML-2016-JiangL #evaluation #learning #robust

Doubly Robust Off-policy Value Evaluation for Reinforcement Learning (NJ, LL0), pp. 652–661.

ICML-2016-QuXO #algorithm #analysis #optimisation #performance #probability

Fast Rate Analysis of Some Stochastic Optimization Algorithms (CQ, HX, CJO), pp. 662–670.

ICML-2016-LiM #nearest neighbour #performance

Fast k-Nearest Neighbour Search via Dynamic Continuous Indexing (KL, JM), pp. 671–679.

ICML-2016-LeKYC #learning #online #predict #sequence

Smooth Imitation Learning for Online Sequence Prediction (HML0, AK, YY, PC0), pp. 680–688.

ICML-2016-ChenKST #community #graph #locality

Community Recovery in Graphs with Locality (YC0, GMK, CS, DT), pp. 689–698.

ICML-2016-ZhuH #optimisation #performance #reduction

Variance Reduction for Faster Non-Convex Optimization (ZAZ, EH), pp. 699–707.

ICML-2016-PatriniNNC #learning #robust

Loss factorization, weakly supervised learning and label noise robustness (GP, FN, RN, MC), pp. 708–717.

ICML-2016-WangMCBPRGUA #analysis #matrix #network

Analysis of Deep Neural Networks with Extended Data Jacobian Matrix (SW, ArM, RC, JAB, MP, MR, KG, GU, ÖA), pp. 718–726.

ICML-2016-ImaizumiH #parametricity

Doubly Decomposing Nonparametric Tensor Regression (MI, KH), pp. 727–736.

ICML-2016-Pedregosa #approximate #optimisation

Hyperparameter optimization with approximate gradient (FP), pp. 737–746.

ICML-2016-Shalev-Shwartz

SDCA without Duality, Regularization, and Individual Convexity (SSS), pp. 747–754.

ICML-2016-AnavaM #sequence

Heteroscedastic Sequences: Beyond Gaussianity (OA, SM), pp. 755–763.

ICML-2016-ZhengTDZ #approach #collaboration

A Neural Autoregressive Approach to Collaborative Filtering (YZ, BT, WD, HZ), pp. 764–773.

ICML-2016-SafranS #network #on the #quality

On the Quality of the Initial Basin in Overspecified Neural Networks (IS, OS), pp. 774–782.

ICML-2016-DunnerFTJ

Primal-Dual Rates and Certificates (CD, SF, MT, MJ), pp. 783–792.

ICML-2016-Shalev-ShwartzW #how #why

Minimizing the Maximal Loss: How and Why (SSS, YW), pp. 793–801.

ICML-2016-Pimentel-Alarcon #clustering #requirements

The Information-Theoretic Requirements of Subspace Clustering with Missing Data (DLPA, RDN), pp. 802–810.

ICML-2016-CohenHK #feedback #graph #learning #online

Online Learning with Feedback Graphs Without the Graphs (AC, TH, TK), pp. 811–819.

ICML-2016-GlaudeP #automaton #learning #probability

PAC learning of Probabilistic Automaton based on the Method of Moments (HG, OP), pp. 820–829.

ICML-2016-MelnykB #modelling

Estimating Structured Vector Autoregressive Models (IM, AB), pp. 830–839.

ICML-2016-Tosh #strict

Mixing Rates for the Alternating Gibbs Sampler over Restricted Boltzmann Machines and Friends (CT), pp. 840–849.

ICML-2016-BlondelIFU #algorithm #network #performance #polynomial

Polynomial Networks and Factorization Machines: New Insights and Efficient Training Algorithms (MB, MI, AF, NU), pp. 850–858.

ICML-2016-GermainHLM #adaptation

A New PAC-Bayesian Perspective on Domain Adaptation (PG, AH, FL, EM), pp. 859–868.

ICML-2016-PuleoM #bound #clustering #correlation #fault

Correlation Clustering and Biclustering with Locally Bounded Errors (GJP, OM), pp. 869–877.

ICML-2016-DavidS #algorithm #bound #performance #problem

PAC Lower Bounds and Efficient Algorithms for The Max K-Armed Bandit Problem (YD, NS), pp. 878–887.

ICML-2016-ElhoseinyEBE #analysis #categorisation #comparative #estimation #modelling #multi

A Comparative Analysis and Study of Multiview CNN Models for Joint Object Categorization and Pose Estimation (ME, TEG, AB, AME), pp. 888–897.

ICML-2016-CarrGL #energy #named #optimisation

BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces (SC, RG, CL), pp. 898–907.

ICML-2016-ArjevaniS #algorithm #complexity #first-order #on the #optimisation

On the Iteration Complexity of Oblivious First-Order Optimization Algorithms (YA, OS), pp. 908–916.

ICML-2016-LiZALH #learning #optimisation #probability

Stochastic Variance Reduced Optimization for Nonconvex Sparse Learning (XL, TZ, RA, HL0, JDH), pp. 917–925.

ICML-2016-Wipf #analysis #estimation #rank

Analysis of Variational Bayesian Factorizations for Sparse and Low-Rank Estimation (DPW), pp. 926–935.

ICML-2016-NewlingF #bound #performance

Fast k-means with accurate bounds (JN, FF), pp. 936–944.

ICML-2016-RavanbakhshPG #matrix #message passing

Boolean Matrix Factorization and Noisy Completion via Message Passing (SR, BP, RG), pp. 945–954.

ICML-2016-CohenS #network

Convolutional Rectifier Networks as Generalized Tensor Decompositions (NC, AS), pp. 955–963.

ICML-2016-TuBSSR #equation #linear #matrix #rank

Low-rank Solutions of Linear Matrix Equations via Procrustes Flow (ST, RB, MS, MS, BR), pp. 964–973.

ICML-2016-JunN #multi #using

Anytime Exploration for Multi-armed Bandits using Confidence Information (KSJ, RDN), pp. 974–982.

ICML-2016-BelangerM #energy #network #predict

Structured Prediction Energy Networks (DB, AM), pp. 983–992.

ICML-2016-ZhangLJ #network #polynomial

L1-regularized Neural Networks are Improperly Learnable in Polynomial Time (YZ0, JDL, MIJ), pp. 993–1001.

ICML-2016-TremblayPGV #clustering

Compressive Spectral Clustering (NT, GP, RG, PV), pp. 1002–1011.

ICML-2016-KasaiM #approach #rank

Low-rank tensor completion: a Riemannian manifold preconditioning approach (HK, BM), pp. 1012–1021.

ICML-2016-ZhangCL #retrieval

Provable Non-convex Phase Retrieval with Outliers: Median TruncatedWirtinger Flow (HZ, YC, YL), pp. 1022–1031.

ICML-2016-DEramoRN #approximate

Estimating Maximum Expected Value through Gaussian Approximation (CD, MR, AN), pp. 1032–1040.

ICML-2016-OswalCRRN #learning #network #similarity

Representational Similarity Learning with Application to Brain Networks (UO, CRC, MALR, TTR, RDN), pp. 1041–1049.

ICML-2016-GalG #approximate #learning #nondeterminism #representation

Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning (YG, ZG), pp. 1050–1059.

ICML-2016-ReedAYLSL #generative #image #synthesis

Generative Adversarial Text to Image Synthesis (SER, ZA, XY, LL, BS, HL), pp. 1060–1069.

ICML-2016-PrabhakaranACP #process

Dirichlet Process Mixture Model for Correcting Technical Variation in Single-Cell Gene Expression Data (SP, EA, AC, DP), pp. 1070–1079.

ICML-2016-ZhuY

Improved SVRG for Non-Strongly-Convex or Sum-of-Non-Convex Objectives (ZAZ, YY), pp. 1080–1089.

ICML-2016-BhowmikGK #parametricity

Sparse Parameter Recovery from Aggregated Data (AB, JG, OK), pp. 1090–1099.

ICML-2016-ZhaiCLZ #detection #energy #modelling

Deep Structured Energy Based Models for Anomaly Detection (SZ, YC, WL, ZZ), pp. 1100–1109.

ICML-2016-ZhuQRY #coordination #performance #using

Even Faster Accelerated Coordinate Descent Using Non-Uniform Sampling (ZAZ, ZQ, PR, YY), pp. 1110–1119.

ICML-2016-ArjovskySB #evolution #network

Unitary Evolution Recurrent Neural Networks (MA, AS, YB), pp. 1120–1128.

ICML-2016-ZhangP #feature model #markov #modelling

Markov Latent Feature Models (AZ, JWP), pp. 1129–1137.

ICML-2016-WangWP #probability

The Knowledge Gradient for Sequential Decision Making with Stochastic Binary Feedbacks (YW, CW, WBP), pp. 1138–1147.

ICML-2016-AsterisKKP #algorithm

A Simple and Provable Algorithm for Sparse Diagonal CCA (MA, AK, OK, RAP), pp. 1148–1157.

ICML-2016-LiuWS #constraints #convergence #linear #optimisation #orthogonal #polynomial

Quadratic Optimization with Orthogonality Constraints: Explicit Lojasiewicz Exponent and Linear Convergence of Line-Search Methods (HL, WW, AMCS), pp. 1158–1167.

ICML-2016-ArpitZKG #network #normalisation #parametricity

Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks (DA, YZ, BUK, VG), pp. 1168–1176.

ICML-2016-LiZZ #learning #memory management

Learning to Generate with Memory (CL, JZ0, BZ0), pp. 1177–1186.

ICML-2016-FernandoG #classification #learning #video

Learning End-to-end Video Classification with Rank-Pooling (BF, SG), pp. 1187–1196.

ICML-2016-SunVBB #learning #predict

Learning to Filter with Predictive State Inference Machines (WS0, AV, BB, JAB), pp. 1197–1205.

ICML-2016-RahmaniA #approach #composition #learning #matrix #performance

A Subspace Learning Approach for High Dimensional Matrix Decomposition with Efficient Column/Row Sampling (MR, GKA), pp. 1206–1214.

ICML-2016-KatariyaKSW #learning #multi #rank

DCM Bandits: Learning to Rank with Multiple Clicks (SK, BK, CS, ZW), pp. 1215–1224.

ICML-2016-HardtRS #performance #probability

Train faster, generalize better: Stability of stochastic gradient descent (MH, BR, YS), pp. 1225–1234.

ICML-2016-KomiyamaHN #algorithm #bound #performance #problem

Copeland Dueling Bandit Problem: Regret Lower Bound, Optimal Algorithm, and Computationally Efficient Algorithm (JK, JH, HN), pp. 1235–1244.

ICML-2016-LiWZC #combinator

Contextual Combinatorial Cascading Bandits (SL0, BW, SZ, WC), pp. 1245–1253.

ICML-2016-WuSLS

Conservative Bandits (YW, RS, TL, CS), pp. 1254–1262.

ICML-2016-HazanL #optimisation #probability

Variance-Reduced and Projection-Free Stochastic Optimization (EH, HL), pp. 1263–1271.

ICML-2016-SongGC #learning #network #sequence

Factored Temporal Sigmoid Belief Networks for Sequence Learning (JS, ZG, LC), pp. 1272–1281.

ICML-2016-XuXCY #assessment #crowdsourcing #quality #ranking #statistics

False Discovery Rate Control and Statistical Quality Assessment of Annotators in Crowdsourced Ranking (QX, JX, XC, YY0), pp. 1282–1291.

ICML-2016-BalduzziG #network

Strongly-Typed Recurrent Neural Networks (DB, MG), pp. 1292–1300.

ICML-2016-KordaSL #clustering #distributed #linear #network

Distributed Clustering of Linear Bandits in Peer to Peer Networks (NK, BS, SL), pp. 1301–1309.

ICML-2016-ZhaoAGA #network

Collapsed Variational Inference for Sum-Product Networks (HZ0, TA, GJG, BA), pp. 1310–1318.

ICML-2016-KhandelwalLNS #analysis #monte carlo #on the

On the Analysis of Complex Backup Strategies in Monte Carlo Tree Search (PK, EL, SN, PS), pp. 1319–1328.

ICML-2016-DuanCHSA #benchmark #learning #metric

Benchmarking Deep Reinforcement Learning for Continuous Control (YD, XC0, RH, JS, PA), pp. 1329–1338.

ICML-2016-DingLHL #clustering #distributed

K-Means Clustering with Distributed Dimensions (HD, YL, LH, JL0), pp. 1339–1348.

ICML-2016-UlyanovLVL #image #network #synthesis

Texture Networks: Feed-forward Synthesis of Textures and Stylized Images (DU, VL, AV, VSL), pp. 1349–1357.

ICML-2016-MirzasoleimanBK #performance #personalisation #summary

Fast Constrained Submodular Maximization: Personalized Data Summarization (BM, AB, AK), pp. 1358–1367.

ICML-2016-WangGL #on the #statistics

On the Statistical Limits of Convex Relaxations (ZW, QG, HL0), pp. 1368–1377.

ICML-2016-KumarIOIBGZPS #memory management #natural language #network

Ask Me Anything: Dynamic Memory Networks for Natural Language Processing (AK, OI, PO, MI, JB0, IG, VZ, RP, RS), pp. 1378–1387.

ICML-2016-ColinBSC #distributed #optimisation

Gossip Dual Averaging for Decentralized Optimization of Pairwise Functions (IC, AB, JS, SC), pp. 1388–1396.

ICML-2016-GonenOS #sketching #using

Solving Ridge Regression using Sketched Preconditioned SVRG (AG, FO, SSS), pp. 1397–1405.

ICML-2016-AJFMS #cumulative #learning #predict

Cumulative Prospect Theory Meets Reinforcement Learning: Prediction and Control (PLA, CJ, MCF0, SIM, CS), pp. 1406–1415.

ICML-2016-PlataniosDM #approach

Estimating Accuracy from Unlabeled Data: A Bayesian Approach (EAP, AD, TMM), pp. 1416–1425.

ICML-2016-BhattacharyaGKP #matrix

Non-negative Matrix Factorization under Heavy Noise (CB, NG, RK, JP), pp. 1426–1434.

ICML-2016-JasinskaDBPKH #probability #using

Extreme F-measure Maximization using Sparse Probability Estimates (KJ, KD, RBF, KP, TK, EH), pp. 1435–1444.

ICML-2016-MaaloeSSW #generative #modelling

Auxiliary Deep Generative Models (LM, CKS, SKS, OW), pp. 1445–1453.

ICML-2016-CanevetJF #empirical #scalability

Importance Sampling Tree for Large-scale Empirical Expectation (OC, CJ, FF), pp. 1454–1462.

ICML-2016-DaneshmandLH #adaptation #learning

Starting Small - Learning with Adaptive Sample Sizes (HD, AL, TH), pp. 1463–1471.

ICML-2016-BuiHHLT #approximate #process #using

Deep Gaussian Processes for Regression using Approximate Expectation Propagation (TDB, DHL, JMHL, YL, RET), pp. 1472–1481.

ICML-2016-MitrovicST #approximate #kernel #named

DR-ABC: Approximate Bayesian Computation with Kernel-Based Distribution Regression (JM, DS, YWT), pp. 1482–1491.

ICML-2016-Hernandez-Lobato #multi #optimisation #predict

Predictive Entropy Search for Multi-objective Bayesian Optimization (DHL, JMHL, AS, RPA), pp. 1492–1501.

ICML-2016-GeZ #analysis #component

Rich Component Analysis (RG0, JZ), pp. 1502–1510.

ICML-2016-Hernandez-Lobato16a #black box

Black-Box Alpha Divergence Minimization (JMHL, YL, MR, TDB, DHL, RET), pp. 1511–1520.

ICML-2016-RezendeMDGW #generative #modelling

One-Shot Generalization in Deep Generative Models (DJR, SM, ID, KG, DW), pp. 1521–1529.

ICML-2016-NatarajanKRD #classification #multi

Optimal Classification with Multivariate Losses (NN, OK, PR, ISD), pp. 1530–1538.

ICML-2016-MalherbeCV #approach #optimisation #ranking

A ranking approach to global optimization (CM, EC, NV), pp. 1539–1547.

ICML-2016-WangSDNSX #algorithm #coordination #distributed #parallel

Parallel and Distributed Block-Coordinate Frank-Wolfe Algorithms (YXW, VS, WD0, WN, SS, EPX), pp. 1548–1557.

ICML-2016-LarsenSLW #encoding #metric #similarity #using

Autoencoding beyond pixels using a learned similarity metric (ABLL, SKS, HL, OW), pp. 1558–1566.

ICML-2016-SaRO #agile #bias

Ensuring Rapid Mixing and Low Bias for Asynchronous Gibbs Sampling (CDS, CR, KO), pp. 1567–1576.

ICML-2016-ShibagakiKHT #modelling

Simultaneous Safe Screening of Features and Samples in Doubly Sparse Modeling (AS, MK, KH, IT), pp. 1577–1586.

ICML-2016-DegenneP #algorithm #multi #probability

Anytime optimal algorithms in stochastic multi-armed bandits (RD, VP), pp. 1587–1595.

ICML-2016-HoilesS #bound #design #education #evaluation #recommendation

Bounded Off-Policy Evaluation with Missing Data for Course Recommendation and Curriculum Design (WH, MvdS), pp. 1596–1604.

ICML-2016-PandeyD #metric #on the #random #representation

On collapsed representation of hierarchical Completely Random Measures (GP0, AD), pp. 1605–1613.

ICML-2016-MartinsA #classification #multi

From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification (AFTM, RFA), pp. 1614–1623.

ICML-2016-BubeckL #black box #optimisation

Black-box Optimization with a Politician (SB, YTL), pp. 1624–1631.

ICML-2016-KanagawaSKST #parametricity #process

Gaussian process nonparametric tensor estimator and its minimax optimality (HK, TS, HK, NS, YT), pp. 1632–1641.

ICML-2016-MedinaY #algorithm #linear

No-Regret Algorithms for Heavy-Tailed Linear Bandits (AMM, SY), pp. 1642–1650.

ICML-2016-BonillaSR

Extended and Unscented Kitchen Sinks (EVB, DMS, AR0), pp. 1651–1659.

ICML-2016-XuZCL #matrix #optimisation #probability

Matrix Eigen-decomposition via Doubly Stochastic Riemannian Optimization (ZX, PZ, JC, XL0), pp. 1660–1669.

ICML-2016-SchnabelSSCJ #evaluation #learning #recommendation

Recommendations as Treatments: Debiasing Learning and Evaluation (TS, AS, AS, NC, TJ), pp. 1670–1679.

ICML-2016-YoonAHS #named #predict

ForecastICU: A Prognostic Decision Support System for Timely Prediction of Intensive Care Unit Admission (JY, AMA, SH, MvdS), pp. 1680–1689.

ICML-2016-LocatelliGC #algorithm #problem

An optimal algorithm for the Thresholding Bandit Problem (AL, MG, AC), pp. 1690–1698.

ICML-2016-NiuRFH #parametricity #performance #using

Fast Parameter Inference in Nonlinear Dynamical Systems using Iterative Gradient Matching (MN, SR, MF, DH), pp. 1699–1707.

ICML-2016-LouizosW #learning #matrix #performance

Structured and Efficient Variational Deep Learning with Matrix Gaussian Posteriors (CL, MW), pp. 1708–1716.

ICML-2016-XuFZ #learning #process

Learning Granger Causality for Hawkes Processes (HX, MF, HZ), pp. 1717–1726.

ICML-2016-MiaoYB

Neural Variational Inference for Text Processing (YM, LY, PB), pp. 1727–1736.

ICML-2016-MenschMTV #learning #matrix #taxonomy

Dictionary Learning for Massive Matrix Factorization (AM, JM, BT, GV), pp. 1737–1746.

ICML-2016-OordKK #network

Pixel Recurrent Neural Networks (AvdO, NK, KK), pp. 1747–1756.

ICML-2016-SimsekAK #problem #why

Why Most Decisions Are Easy in Tetris - And Perhaps in Other Sequential Decision Problems, As Well (ÖS, SA, AK), pp. 1757–1765.

ICML-2016-LiSJ #matrix

Gaussian quadrature for matrix inverse forms with applications (CL, SS, SJ), pp. 1766–1775.

ICML-2016-MeshiMWS #predict

Train and Test Tightness of LP Relaxations in Structured Prediction (OM, MM, AW, DAS), pp. 1776–1785.

ICML-2016-AroraMM #learning #multi #optimisation #probability #representation #using

Stochastic Optimization for Multiview Representation Learning using Partial Least Squares (RA, PM, TVM), pp. 1786–1794.

ICML-2016-BasbugE

Hierarchical Compound Poisson Factorization (MEB, BEE), pp. 1795–1803.

ICML-2016-HeB #learning #modelling

Opponent Modeling in Deep Reinforcement Learning (HH0, JLBG), pp. 1804–1813.

ICML-2016-WangDL #linear #modelling

No penalty no tears: Least squares in high-dimensional linear models (XW0, DBD, CL), pp. 1814–1822.

ICML-2016-QuRTF #empirical #named #probability

SDNA: Stochastic Dual Newton Ascent for Empirical Risk Minimization (ZQ, PR, MT, OF), pp. 1823–1832.

ICML-2016-HazanLS #on the #optimisation #probability #problem

On Graduated Optimization for Stochastic Non-Convex Problems (EH, KYL, SSS), pp. 1833–1841.

ICML-2016-SantoroBBWL #network

Meta-Learning with Memory-Augmented Neural Networks (AS, SB, MB, DW, TPL), pp. 1842–1850.

ICML-2016-DaiB #multi

The knockoff filter for FDR control in group-sparse and multitask regression (RD, RB), pp. 1851–1859.

ICML-2016-PerolatPGSP #approximate #game studies #markov #policy

Softened Approximate Policy Iteration for Markov Games (JP, BP, MG, BS, OP), pp. 1860–1868.

ICML-2016-GowerGR #probability

Stochastic Block BFGS: Squeezing More Curvature out of Data (RMG, DG, PR), pp. 1869–1878.

ICML-2016-BaiRWS #classification #difference #geometry #learning

Differential Geometric Regularization for Supervised Learning of Classifiers (QB, SR, ZW, SS), pp. 1879–1888.

ICML-2016-DielemanFK #network #symmetry

Exploiting Cyclic Symmetry in Convolutional Neural Networks (SD, JDF, KK), pp. 1889–1898.

ICML-2016-ZahavyBM #black box #comprehension

Graying the black box: Understanding DQNs (TZ, NBZ, SM), pp. 1899–1908.

ICML-2016-FriesenD #learning #modelling #theorem

The Sum-Product Theorem: A Foundation for Learning Tractable Models (ALF, PMD), pp. 1909–1918.

ICML-2016-ShahG #correlation #learning

Pareto Frontier Learning with Expensive Correlated Objectives (AS, ZG), pp. 1919–1927.

ICML-2016-MnihBMGLHSK #learning

Asynchronous Methods for Deep Reinforcement Learning (VM, APB, MM, AG, TPL, TH, DS, KK), pp. 1928–1937.

ICML-2016-VeldtGM

A Simple and Strongly-Local Flow-Based Method for Cut Improvement (NV, DFG, MWM), pp. 1938–1947.

ICML-2016-SuLCC #learning #modelling #statistics #visual notation

Nonlinear Statistical Learning with Truncated Gaussian Graphical Models (QS, XL, CC, LC), pp. 1948–1957.

ICML-2016-KawakitaT #learning

Barron and Cover's Theory in Supervised Learning and its Application to Lasso (MK, JT), pp. 1958–1966.

ICML-2016-MichaeliWL #analysis #canonical #correlation #parametricity

Nonparametric Canonical Correlation Analysis (TM, WW, KL), pp. 1967–1976.

ICML-2016-RakhlinS #named #performance

BISTRO: An Efficient Relaxation-Based Method for Contextual Bandits (AR, KS), pp. 1977–1985.

ICML-2016-DanihelkaWUKG #memory management

Associative Long Short-Term Memory (ID, GW, BU, NK, AG), pp. 1986–1994.

ICML-2016-WangSHHLF #architecture #learning #network

Dueling Network Architectures for Deep Reinforcement Learning (ZW0, TS, MH, HvH, ML, NdF), pp. 1995–2003.

ICML-2016-KusanoHF #data analysis #kernel #persistent

Persistence weighted Gaussian kernel for topological data analysis (GK, YH, KF), pp. 2004–2013.

ICML-2016-NiepertAK #graph #learning #network

Learning Convolutional Neural Networks for Graphs (MN, MA, KK), pp. 2014–2023.

ICML-2016-DiamosSCCCEEHS #persistent

Persistent RNNs: Stashing Recurrent Weights On-Chip (GD, SS, BC, MC, AC, EE, JHE, AYH, SS), pp. 2024–2033.

ICML-2016-HenaffSL #network #orthogonal

Recurrent Orthogonal Networks and Long-Memory Tasks (MH, AS, YL), pp. 2034–2042.

ICML-2016-BauerSP #multi

The Arrow of Time in Multivariate Time Series (SB, BS, JP), pp. 2043–2051.

ICML-2016-RamaswamyST #estimation #kernel

Mixture Proportion Estimation via Kernel Embeddings of Distributions (HGR, CS, AT), pp. 2052–2060.

ICML-2016-LiJS #kernel #performance

Fast DPP Sampling for Nystrom with Application to Kernel Methods (CL, SJ, SS), pp. 2061–2070.

ICML-2016-TrouillonWRGB #predict

Complex Embeddings for Simple Link Prediction (TT, JW, SR0, ÉG, GB), pp. 2071–2080.

ICML-2016-VikramD #clustering #interactive

Interactive Bayesian Hierarchical Clustering (SV, SD), pp. 2081–2090.

ICML-2016-AllamanisPS #network #source code #summary

A Convolutional Attention Network for Extreme Summarization of Source Code (MA, HP, CAS), pp. 2091–2100.

ICML-2016-KapralovPW #how #matrix #multi

How to Fake Multiply by a Gaussian Matrix (MK, VKP, DPW), pp. 2101–2110.

ICML-2016-RogersVLG #independence #testing

Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing (MG, HWL, RMR, SPV), pp. 2111–2120.

ICML-2016-ErraqabiVCM

Pliable Rejection Sampling (AE, MV, AC, OAM), pp. 2121–2129.

ICML-2016-BalleGP #evaluation #policy

Differentially Private Policy Evaluation (BB, MG, DP), pp. 2130–2138.

ICML-2016-ThomasB #evaluation #learning #policy

Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning (PST, EB), pp. 2139–2148.

ICML-2016-WiatowskiTSGB #architecture #feature model

Discrete Deep Feature Extraction: A Theory and New Architectures (TW, MT, AS, PG, HB), pp. 2149–2158.

ICML-2016-SyrgkanisKS #algorithm #learning #performance

Efficient Algorithms for Adversarial Contextual Learning (VS, AK, RES), pp. 2159–2168.

ICML-2016-SongSZU #network

Training Deep Neural Networks via Direct Loss Minimization (YS, AGS, RSZ, RU), pp. 2169–2177.

ICML-2016-HwangS #sequence

Sequence to Sequence Training of CTC-RNNs with Partial Windowing (KH, WS), pp. 2178–2187.

ICML-2016-MnihR #monte carlo

Variational Inference for Monte Carlo Objectives (AM, DJR), pp. 2188–2196.

ICML-2016-DalalGM #grid

Hierarchical Decision Making In Electricity Grid Management (GD, EG, SM), pp. 2197–2206.

ICML-2016-BalkanskiMKS #combinator #learning

Learning Sparse Combinatorial Representations via Two-stage Submodular Maximization (EB, BM, AK0, YS), pp. 2207–2216.

ICML-2016-ShangSAL #comprehension #linear #network

Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units (WS, KS, DA, HL), pp. 2217–2225.

ICML-2016-WangXDS #process

Isotonic Hawkes Processes (YW0, BX0, ND, LS), pp. 2226–2234.

ICML-2016-LiuY #learning #multi

Cross-Graph Learning of Multi-Relational Associations (HL, YY), pp. 2235–2243.

ICML-2016-PanRAG #process

Markov-modulated Marked Poisson Processes for Check-in Data (JP, VAR, PKA, AEG), pp. 2244–2253.

ICML-2016-AchimSE #analysis #constraints #fourier

Beyond Parity Constraints: Fourier Analysis of Hash Functions for Inference (TA, AS, SE), pp. 2254–2262.

ICML-2016-Papakonstantinou #learning #on the

On the Power and Limits of Distance-Based Learning (PAP, JX0, GY), pp. 2263–2271.

ICML-2016-YenLZRD #approach #multi #sequence

A Convex Atomic-Norm Approach to Multiple Sequence Alignment and Motif Discovery (IEHY, XL, JZ, PR, ISD), pp. 2272–2280.

ICML-2016-FazayeliB #estimation

Generalized Direct Change Estimation in Ising Model Structure (FF, AB), pp. 2281–2290.

ICML-2016-ChiangHD #analysis #component #robust

Robust Principal Component Analysis with Side Information (KYC, CJH, ISD), pp. 2291–2299.

ICML-2016-GuiHG #estimation #matrix #performance #rank #towards

Towards Faster Rates and Oracle Property for Low-Rank Matrix Estimation (HG, JH0, QG), pp. 2300–2309.

ICML-2016-SangnierGR #detection #proximity #reliability #representation #using

Early and Reliable Event Detection Using Proximity Space Representation (MS, JG, AR), pp. 2310–2319.

ICML-2016-LibertyLS #machine learning

Stratified Sampling Meets Machine Learning (EL, KJL, KS), pp. 2320–2329.

ICML-2016-GuanRW #learning #markov #multi #performance #process #recognition #using

Efficient Multi-Instance Learning for Activity Recognition from Time Series Data Using an Auto-Regressive Hidden Markov Model (XG, RR, WKW), pp. 2330–2339.

ICML-2016-LinCR #multi

Generalization Properties and Implicit Regularization for Multiple Passes SGM (JL, RC, LR), pp. 2340–2348.

ICML-2016-FrostigMMS #analysis #component

Principal Component Projection Without Principal Component Analysis (RF, CM, CM, AS), pp. 2349–2357.

ICML-2016-LiLR #approximate #rank

Recovery guarantee of weighted low-rank approximation via alternating minimization (YL, YL, AR), pp. 2358–2367.

ICML-2016-PezeshkiFBCB #architecture #network

Deconstructing the Ladder Network Architecture (MP, LF, PB, ACC, YB), pp. 2368–2376.

ICML-2016-OsbandRW #random

Generalization and Exploration via Randomized Value Functions (IO, BVR, ZW), pp. 2377–2386.

ICML-2016-KantchelianTJ #classification

Evasion and Hardening of Tree Ensemble Classifiers (AK, JDT, ADJ), pp. 2387–2396.

ICML-2016-XiongMS #memory management #network #visual notation

Dynamic Memory Networks for Visual and Textual Question Answering (CX, SM, RS), pp. 2397–2406.

ICML-2016-RavanbakhshOFPH #matter #parametricity

Estimating Cosmological Parameters from the Dark Matter Distribution (SR, JBO, SF, LP, SH, JGS, BP), pp. 2407–2416.

ICML-2016-HashimotoGJ #generative #learning

Learning Population-Level Diffusions with Generative RNNs (TBH, DKG, TSJ), pp. 2417–2426.

ICML-2016-PanS #network

Expressiveness of Rectifier Networks (XP, VS), pp. 2427–2435.

ICML-2016-KairouzBR #estimation #privacy

Discrete Distribution Estimation under Local Privacy (PK, KB, DR), pp. 2436–2444.

ICML-2016-InouyeRD #dependence #exponential #modelling #multi #product line #visual notation

Square Root Graphical Models: Multivariate Generalizations of Univariate Exponential Families that Permit Positive Dependencies (DII, PR, ISD), pp. 2445–2453.

ICML-2016-LimW #approach #permutation #problem

A Box-Constrained Approach for Hard Permutation Problems (CHL, SW), pp. 2454–2463.

ICML-2016-ZadehHS #geometry #learning #metric

Geometric Mean Metric Learning (PZ, RH, SS), pp. 2464–2471.

ICML-2016-YangWLEZ #estimation #parametricity

Sparse Nonlinear Regression: Parameter Estimation under Nonconvexity (ZY, ZW, HL0, YCE, TZ0), pp. 2472–2481.

ICML-2016-LiWPA #classification #multi

Conditional Bernoulli Mixtures for Multi-label Classification (CL, BW, VP, JAA), pp. 2482–2491.

ICML-2016-ChenG #multi #problem #scalability

Scalable Discrete Sampling as a Multi-Armed Bandit Problem (YC, ZG), pp. 2492–2501.

ICML-2016-ChoromanskiS #kernel #sublinear

Recycling Randomness with Structure for Sublinear time Kernel Expansions (KC, VS), pp. 2502–2510.

ICML-2016-BornscheinSFB #bidirectional

Bidirectional Helmholtz Machines (JB, SS, AF, YB), pp. 2511–2519.

ICML-2016-AbernethyH #optimisation #performance

Faster Convex Optimization: Simulated Annealing with an Efficient Universal Barrier (JDA, EH), pp. 2520–2528.

ICML-2016-CutajarOCF #kernel #matrix

Preconditioning Kernel Matrices (KC, MAO, JPC, MF), pp. 2529–2538.

ICML-2016-AltschulerBFMRZ #algorithm #bound #distributed #set

Greedy Column Subset Selection: New Bounds and Distributed Algorithms (JA, AB, GF, VSM, AR, MZ), pp. 2539–2548.

ICML-2016-AlmahairiBCZLC #capacity #network

Dynamic Capacity Networks (AA, NB, TC, YZ, HL, ACC), pp. 2549–2558.

ICML-2016-HeidariMSVY

Pricing a Low-regret Seller (HH, MM, US, SV, SY), pp. 2559–2567.

ICML-2016-RaghunathanFDL #estimation #linear

Estimation from Indirect Supervision with Linear Moments (AR, RF, JCD, PL), pp. 2568–2577.

ICML-2016-BotteschBK #approximate

Speeding up k-means by approximating Euclidean distances via block vectors (TB, TB, MK), pp. 2578–2586.

ICML-2016-MussmannE #learning

Learning and Inference via Maximum Inner Product Search (SM, SE), pp. 2587–2596.

ICML-2016-RodomanovK #finite #optimisation

A Superlinearly-Convergent Proximal Newton-type Method for the Optimization of Finite Sums (AR, DK), pp. 2597–2605.

ICML-2016-ChwialkowskiSG #kernel

A Kernel Test of Goodness of Fit (KC, HS, AG), pp. 2606–2615.

ICML-2016-RainforthNLPMDW #markov #monte carlo

Interacting Particle Markov Chain Monte Carlo (TR, CAN, FL, BP, JWvdM, AD, FDW), pp. 2616–2625.

ICML-2016-GarberHJKMNS #performance

Faster Eigenvector Computation via Shift-and-Invert Preconditioning (DG, EH, CJ, SMK, CM, PN, AS), pp. 2626–2634.

ICML-2016-XieLZW #formal method #generative

A Theory of Generative ConvNet (JX, YL0, SCZ, YNW), pp. 2635–2644.

ICML-2016-YaoK #learning #performance #product line

Efficient Learning with a Family of Nonconvex Regularizers by Redistributing Nonconvexity (QY, JTK), pp. 2645–2654.

ICML-2016-SiHD #approximate #performance #using

Computationally Efficient Nyström Approximation using Fast Transforms (SS, CJH, ISD), pp. 2655–2663.

ICML-2016-PeyreCS #distance #kernel #matrix

Gromov-Wasserstein Averaging of Kernel and Distance Matrices (GP, MC, JS), pp. 2664–2672.

ICML-2016-RoychowdhuryKP #monte carlo #robust #using

Robust Monte Carlo Sampling using Riemannian Nosé-Poincaré Hamiltonian Dynamics (AR, BK, SP0), pp. 2673–2681.

ICML-2016-SaeediHJA #infinity #performance

The Segmented iHMM: A Simple, Efficient Hierarchical Infinite HMM (AS, MDH, MJJ0, RPA), pp. 2682–2691.

ICML-2016-UstinovskiyFGS #learning

Meta-Gradient Boosted Decision Tree Model for Weight and Target Learning (YU, VF, GG, PS), pp. 2692–2701.

ICML-2016-DaiDS #modelling

Discriminative Embeddings of Latent Variable Models for Structured Data (HD, BD, LS), pp. 2702–2711.

ICML-2016-GuhaMRS #detection #random #robust

Robust Random Cut Forest Based Anomaly Detection on Streams (SG, NM, GR, OS), pp. 2712–2721.

ICML-2016-TaylorBXSPG #approach #network #scalability

Training Neural Networks Without Gradients: A Scalable ADMM Approach (GT, RB, ZX0, BS, ABP, TG), pp. 2722–2731.

ICML-2016-ChenQ #category theory #clustering

Clustering High Dimensional Categorical Data via Topographical Features (CC0, NQ), pp. 2732–2740.

ICML-2016-GeJKNS #algorithm #analysis #canonical #correlation #performance #scalability

Efficient Algorithms for Large-scale Generalized Eigenvector Computation and Canonical Correlation Analysis (RG0, CJ, SMK, PN, AS), pp. 2741–2750.

ICML-2016-BaiIWB #algorithm #optimisation

Algorithms for Optimizing the Ratio of Submodular Functions (WB, RKI, KW, JAB), pp. 2751–2759.

ICML-2016-HoGE #learning #optimisation #policy

Model-Free Imitation Learning with Policy Optimization (JH, JKG, SE), pp. 2760–2769.

ICML-2016-CisseAB #architecture #named

ADIOS: Architectures Deep In Output Space (MC, MAS, SB), pp. 2770–2779.

ICML-2016-GaoKOV #axiom #capacity #dependence

Conditional Dependence via Shannon Capacity: Axioms, Estimators and Applications (WG, SK, SO, PV), pp. 2780–2789.

ICML-2016-OhCSL #memory management

Control of Memory, Active Perception, and Action in Minecraft (JO, VC, SPS, HL), pp. 2790–2799.

ICML-2016-SuhZA #classification #complexity

The Label Complexity of Mixed-Initiative Classifier Training (JS, XZ0, SA), pp. 2800–2809.

ICML-2016-ScheinZBW #composition #learning

Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations (AS, MZ, DMB, HMW), pp. 2810–2819.

ICML-2016-ColomboV #composition #matrix

Tensor Decomposition via Joint Matrix Schur Decomposition (NC, NV), pp. 2820–2828.

ICML-2016-GuLSL #modelling

Continuous Deep Q-Learning with Model-based Acceleration (SG, TPL, IS, SL), pp. 2829–2838.

ICML-2016-GongZLTGS #adaptation #component

Domain Adaptation with Conditional Transferable Components (MG, KZ0, TL, DT, CG, BS), pp. 2839–2848.

ICML-2016-LinTA #fixpoint #network

Fixed Point Quantization of Deep Convolutional Networks (DDL, SST, VSA), pp. 2849–2858.

ICML-2016-AroraGKMM #algorithm #modelling #topic

Provable Algorithms for Inference in Topic Models (SA, RG0, FK, TM, AM), pp. 2859–2867.

ICML-2016-WangWK #performance #programming

Epigraph projections for fast general convex programming (PWW, MW, JZK), pp. 2868–2877.

ICML-2016-AcharyaDLS #algorithm #performance

Fast Algorithms for Segmented Regression (JA, ID, JL0, LS), pp. 2878–2886.

ICML-2016-ThomasSDB

Energetic Natural Gradient Descent (PST, BCdS, CD, EB), pp. 2887–2895.

ICML-2016-CarlsonSPP

Partition Functions from Rao-Blackwellized Tempered Sampling (DEC, PS, AP, LP), pp. 2896–2905.

ICML-2016-ZhaoPX #learning #modelling

Learning Mixtures of Plackett-Luce Models (ZZ, PP, LX), pp. 2906–2914.

ICML-2016-AbelHL #abstraction #approximate #behaviour

Near Optimal Behavior via Approximate State Abstraction (DA, DEH, MLL), pp. 2915–2923.

ICML-2016-LeiF #order #power of #testing

Power of Ordered Hypothesis Testing (LL, WF), pp. 2924–2932.

ICML-2016-BielikRV #named #probability

PHOG: Probabilistic Model for Code (PB, VR, MTV), pp. 2933–2942.

ICML-2016-GyorgyS #matrix

Shifting Regret, Mirror Descent, and Matrices (AG, CS), pp. 2943–2951.

ICML-2016-LuketinaRBG #scalability

Scalable Gradient-Based Tuning of Continuous Regularization Hyperparameters (JL, TR, MB, KG), pp. 2952–2960.

ICML-2016-AkrourNAA #learning #optimisation

Model-Free Trajectory Optimization for Reinforcement Learning (RA, GN, HA, AA), pp. 2961–2970.

ICML-2016-JiaoKS #distance

Controlling the distance to a Kemeny consensus without computing it (YJ, AK, ES), pp. 2971–2980.

ICML-2016-LucicBZK #scalability

Horizontally Scalable Submodular Maximization (ML, OB, MZ, AK0), pp. 2981–2989.

ICML-2016-CohenW #network

Group Equivariant Convolutional Networks (TC, MW), pp. 2990–2999.

ICML-2016-PiatkowskiM #probability

Stochastic Discrete Clenshaw-Curtis Quadrature (NP, KM), pp. 3000–3009.

ICML-2016-RiemerVCHHK #multi

Correcting Forecasts with Multifactor Neural Attention (MR, AV, FdPC, FFTHI, RH0, EK), pp. 3010–3019.

ICML-2016-JohanssonSS #learning

Learning Representations for Counterfactual Inference (FDJ, US, DAS), pp. 3020–3029.

ICML-2016-HwangTC #automation #modelling #multi #parametricity #relational

Automatic Construction of Nonparametric Relational Regression Models for Multiple Time Series (YH, AT, JC), pp. 3030–3039.

ICML-2016-PaigeW #modelling #monte carlo #network #visual notation

Inference Networks for Sequential Monte Carlo in Graphical Models (BP, FDW), pp. 3040–3049.

ICML-2016-Bloem-ReddyC #slicing

Slice Sampling on Hamiltonian Trajectories (BBR, JC), pp. 3050–3058.

ICML-2016-GulcehreMDB

Noisy Activation Functions (ÇG, MM, MD, YB), pp. 3059–3068.

ICML-2016-YenHRZD #approach #classification #multi

PD-Sparse : A Primal and Dual Sparse Approach to Extreme Multiclass and Multilabel Classification (IEHY, XH, PR, KZ, ISD), pp. 3069–3077.