8:00am EST
8:15am EST
8:30am EST
Introduction 513 ab, ApproxBayesInf
Opening remarks, winner ISBA travel award 511 a, ABC
Opening Remarks 512 cg, CognitiveComputation1
Introduction to the workshop 512 bf, Metareasoning
Introduction to the 'Flavours of Physics' challenge. Andrey Ustyuzhanin (Yandex) 515 bc, ALEPH
Surya Ganguli, Towards a theory of high dimensional, single trial neural data analysis: On the role of random projections and phase transitions 511 f, StatsNeuralSys
Jorge Nocedal, An Evolving Gradient Resampling Method 510 ac, OPT
8:35am EST
8:40am EST
8:45am EST
8:50am EST
9:00am EST
Opening Remarks 512 a, ProbIntegration
Opening Remarks 514 bc, TimeSeries
Theophane Weber: Reinforced Variational Inference (contributed) 513 ab, ApproxBayesInf
Introduction, Multimodal Machine Learning: A Short Survey 512 dh, Multimodal
Data Science Competition Platforms. Ben Hamner (Kaggle) 515 bc, ALEPH
Pedro Domingos 512 cg, CognitiveComputation1
Bento: Learning Stochastic Differential Equations – Fundamental limits and efficient algorithms 511 e, ComplexNetworks
Honglak Lee 513 cd, DeepRL
Introduction by organizers 511 d, FasterFromEasy
Yee Whye Teh (Random Tensor Decompositions for Regression and Collaborative Filtering) 511 c, NonParam
Katherine Heller, Translating between human & animal studies via Bayesian multi-task learning 511 f, StatsNeuralSys
Supervised learning labels in a fast moving environment, Alessandro Magnani (@WalmartsLab) 512 e, MLEcom
Gabriel Kreiman - Visual pattern completion: from neural circuits to computational models 515 a, Neuroimaging1
Keynote (Jason Weston, Facebook AI) 511 b, SpokenLanguage
9:10am EST
9:15am EST
9:20am EST
9:30am EST
9:40am EST
9:45am EST
9:50am EST
10:00am EST
10:20am EST
10:30am EST
Flavours of Physics challenge: 3d place solution. Josef Slavicek 515 bc, ALEPH
Rick Lewis 512 bf, Metareasoning
Andrea Montanari: Approximate inference with semidefinite relaxations 513 ab, ApproxBayesInf
Cynthia Dwork 514 a, AdaptiveDA
Rina Dechter 512 cg, CognitiveComputation1
Macke: Correlations and signatures of criticality in neural population models 511 e, ComplexNetworks
Aarti Singh 511 d, FasterFromEasy
Fei Sha (Do shallow kernel methods match deep neural networks -- and if not, what can the shallow ones learn from the deep ones?) 511 c, NonParam
Arthur Gretton 512 a, ProbIntegration
Poster Session 1 511 f, StatsNeuralSys
David Nott, Uses of ABC in prior choice and Bayesian model checking 511 a, ABC
Michael Mahoney, Column Subset Selection on Terabyte-sized Scientific Data 513 ef, FeatureEx
Why would you recommend me THAT!?, Aish Fenton (Netflix) 512 e, MLEcom
Accepted orals and spotlights 512 dh, Multimodal
Panel discussion: Modern Challenges in Time Series Analysis 514 bc, TimeSeries
10:40am EST
10:50am EST
10:55am EST
11:00am EST
Physics Prize Awards Announcement. Marcin Chrzaszcz (Universitaet Zuerich) 515 bc, ALEPH
Henry Brighton 512 bf, Metareasoning
Accuracy on the test set is not enough --- the risk of deploying unintelligible models in healthcare. Rich Caruana 510 db, MLHC
Jon Ullman 'Barriers to Preventing False Discovery in Interactive Data Analysis' 514 a, AdaptiveDA
Josh Tenenbaum 512 cg, CognitiveComputation1
Poster spotlights part II 511 e, ComplexNetworks
Vlad Mnih 513 cd, DeepRL
Dylan Foster 511 d, FasterFromEasy
Poster highlights 515 a, Neuroimaging1
Poster Spotlights 511 c, NonParam
Roman Garnett 512 a, ProbIntegration
Matthias Bethge, Let's compete - benchmarking models in neuroscience 511 f, StatsNeuralSys
11:10am EST
11:15am EST
11:20am EST
11:30am EST
Dean Foster, Discussant 514 a, AdaptiveDA
Attribute Extraction from Noisy Text Using Character-based Sequence Tagging Models, Pallika Kanani 512 e, MLEcom
Falk Lieder 512 bf, Metareasoning
Discussion, Q&A, etc. 512 cg, CognitiveComputation1
Poster session part I 511 e, ComplexNetworks
Gerry Tesauro 513 cd, DeepRL
Poster spotlights I 511 d, FasterFromEasy
Healthcare challenges #1 510 db, MLHC
Poster session 1 515 a, Neuroimaging1
Posters 511 c, NonParam
Spotlight talks 512 a, ProbIntegration
Yoshua Bengio, Small Steps Towards Biologically Plausible Deep Learning 511 f, StatsNeuralSys
Poster session 513 ab, ApproxBayesInf
11:35am EST
11:45am EST
11:50am EST
12:00pm EST
12:05pm EST
12:10pm EST
12:20pm EST
12:45pm EST
1:15pm EST
1:30pm EST
1:45pm EST
2:00pm EST
2:15pm EST
2:30pm EST
Veeranjaneyulu Sadhanala (Graph Sparsification Approaches for Laplacian Smoothing, Contributed) 511 c, NonParam
Quentin Huys 512 bf, Metareasoning
Montanari: Information-theoretic bounds on learning network dynamics 511 e, ComplexNetworks
Yoshua Bengio 513 cd, DeepRL
Peter Grünwald 511 d, FasterFromEasy
Optimal A-B Testing, Vivek Farias (MIT) 512 e, MLEcom
Francis Bach 512 a, ProbIntegration
Pulkit Agrawal The Human Visual Hierarchy is Isomorphic to the Hierarchy learned by a Deep Convolutional Neural Network Trained for Object Recognition 511 f, StatsNeuralSys
Rob Deardon, ABC-based inference for epidemic models with uncertain underlying contact networks 511 a, ABC
Emily Fox, University of Washington (invited) 514 bc, TimeSeries
Raymond Mooney (UT Austin), Generating Natural-Language Video Descriptions using LSTM Recurrent Neural Networks 512 dh, Multimodal
Mitsuo Kawato - Spectrum of Psychiatric Disorders revealed by Machine Learning Algorithms 515 a, Neuroimaging1
Guanghui Lan, Complexity of composite optimization 510 ac, OPT
2:40pm EST
2:45pm EST
2:50pm EST
3:00pm EST
Angela Yu 512 bf, Metareasoning
Braunstein: Bayesian inference of cascades on networks 511 e, ComplexNetworks
Satyen Kale 511 d, FasterFromEasy
Real-time Predictions using Time-series Data, Devavrat Shah (MIT) 512 e, MLEcom
Jean-Philippe Vert (Learning from Rankings) 511 c, NonParam
David Duvenaud 512 a, ProbIntegration
Yann Lecun, Unsupervised Learning (TBA) 511 f, StatsNeuralSys
Spotlight talks 513 cd, DeepRL
Best 3 paper talks 511 b, SpokenLanguage
3:05pm EST
3:10pm EST
3:15pm EST
3:20pm EST
3:25pm EST
3:30pm EST
Jet Images: Deep Learning Edition. Luke Percival De Oliveira (SLAC National Accelerator Lab.) 515 bc, ALEPH
A Ranking Approach to Address the Click Sparsity Problem in Personalized Ad Recommendation, Sougata Chaudhuri 512 e, MLEcom
Antoine Bordes 512 cg, CognitiveComputation1
Poster session part II 511 e, ComplexNetworks
Poster spotlights II 511 d, FasterFromEasy
Contributed - Romy Lorenz - Stopping criteria for boosting automatic experimental design using real-time fMRI with Bayesian optimization 515 a, Neuroimaging1
Michael Mahoney (Using Local Spectral Methods in Theory and in Practice) 511 c, NonParam
Max Welling 512 a, ProbIntegration
Poster Session 2 511 f, StatsNeuralSys
POSTER SESSION 510 ac, OPT
3:35pm EST
3:40pm EST
3:45pm EST
4:00pm EST
4:20pm EST
4:30pm EST
Stochastically Transitive Models for Pairwise Comparisons: Statistical and Computational Issues, Nihar Shah 512 e, MLEcom
Manfred Opper: Approximate inference for Ising models with random couplings 513 ab, ApproxBayesInf
Wentao Li, On the Asymptotic Behavior of ABC 511 a, ABC
Andrew Gelman 514 a, AdaptiveDA
Poster session part III 511 e, ComplexNetworks
Francis Bach (Sharp Analysis of Random Feature Expansions) 511 c, NonParam
Neil Lawrence, The Mechanistic Fallacy and Modelling how we Think 511 f, StatsNeuralSys
An alternative to ABC for likelihood-free inference. Kyle Stuart Cranmer (NYU) 515 bc, ALEPH
Panel Discussion 513 ef, FeatureEx
Shie Mannor, Technion (invited) 514 bc, TimeSeries
Ruslan Salakhutdinov (CMU), Generating Images from Captions with Attention 512 dh, Multimodal
Poster Session 512 bf, Metareasoning
Sylvain Baillet - Possible mechanisms enabling functional brain connectivity in the resting and active states 515 a, Neuroimaging1
Panel discussion 512 a, ProbIntegration
4:45pm EST
4:55pm EST
5:00pm EST
Automatic Layout Element Detection From E-Commerce Pages, Anura Bhardwaj. 512 e, MLEcom
Closing Remarks 511 c, NonParam
Greg Wayne 512 cg, CognitiveComputation1
Grima: Exact and approximate solutions for spatial stochastic models of chemical system 511 e, ComplexNetworks
Martin Riedmiller 513 cd, DeepRL
Iain Murray, ABC as Learning 511 a, ABC
Gergely Neu 511 d, FasterFromEasy
Panel 514 a, AdaptiveDA
Panel: Deep Learning and neuroscience: What can brains tell us about massive computing and vice versa? 511 f, StatsNeuralSys
5:10pm EST
5:15pm EST
5:25pm EST
5:30pm EST
Deep Temporal Features to Predict Repeat Buyers, Pankaj Malhotra 512 e, MLEcom
Discussion, Q&A, etc. 512 cg, CognitiveComputation1
Taylor: Learning Multi-scale Temporal Dynamics with Recurrent Neural Networks 511 e, ComplexNetworks
Jan Koutnik 513 cd, DeepRL
Personalized Mobile Health Interventions Ambuj Tewari 510 db, MLHC
Panel Discussion 512 bf, Metareasoning
Panel discussion 1 + snacks and drinks 515 a, Neuroimaging1
Elad Hazan, Faster Convex Optimization: Simulated Annealing with an Efficient Universal Barrier 510 ac, OPT
5:45pm EST
5:50pm EST
6:00pm EST
6:05pm EST
6:15pm EST
8:00am EST
8:10am EST
8:20am EST
8:25am EST
8:30am EST
8:40am EST
8:45am EST
8:50am EST
8:55am EST
9:00am EST
Welcome and introduction. Evelyne Viegas 512 e, CiML
Welcome & Open Remarks 512 dh, Privacy
Opening Remarks 512 bf, NetworksInScience
Zoubin Ghahramani 511b, BayesOpt
Jeff Miller 515 bc, BayesNonparam
Methods overview: Andreas Schaefer, University College London 511 e, BigNeuro
Sourek: Lifted Relational Neural Networks 512 cg, CognitiveComputation2
Nathan Wiebe (Intro to quantum computing) 512 a, QuantumML
Trevor Darrell 514 bc, TransferAndMultitask
Random sampling of bandlimited signals on graphs, Pierre Vandergheynst 511 c, Multiresolution
Adventures on the efficient frontier by Andrew Gelman (Columbia University) 513 ab, ScalableMC
Brendan Frey. Learning deep biological architectures for genomic medicine. 510 db, MLCB
Michael Littman, Brown University: 'Reinforcement Learning from users: New algorithms and frameworks' 514 a, Personalization
John Anderson - The Sequential Structure of Thought 515 a, Neuroimaging2
9:05am EST
9:10am EST
9:15am EST
9:20am EST
9:30am EST
9:40am EST
9:45am EST
10:00am EST
10:05am EST
10:10am EST
10:20am EST
10:30am EST
David Kelley. Basset: Learning the regulatory code of the accessible genome with deep convolutional neural networks 510 db, MLCB
Frank Hutter 511b, BayesOpt
Koray Kavukcuoglu 513 ef, BlackBox
Mike Hughes 515 bc, BayesNonparam
Daniel L. Silver 512 cg, CognitiveComputation2
Marius Kloft (HU Berlin) Multi-class SVMs: From Tighter Data-Dependent Generalization Bounds to Novel Algorithms 511 f, Extreme
Beyond Nodes and Edges: Multiresolution Models of Complex Networks, Jure Leskovec 511 c, Multiresolution
Greg Valiant 513 cd, NonConvexOpt
Mehryar Mohri 514 bc, TransferAndMultitask
Alex Graves, Google Deepmind (Invited) 510 ac, RAM
Michael L. Littman (Brown University) 511 a, MultiAgent
Invited Speaker: Nando de Freitas 512 dh, Privacy
Ashish Kapoor (Quantum Deep Learning) 512 a, QuantumML
Poster spotlight session 512 bf, NetworksInScience
Jerry Zhu, University of Wisconsin Madison: 'Machine Teaching for Personalized Education, Security, Interactive Machine Learning' 514 a, Personalization
10:50am EST
10:55am EST
11:00am EST
11:05am EST
11:10am EST
11:15am EST
11:20am EST
11:30am EST
Mohammad Javad Hosseini. Learning Gaussian Graphical Models with Overlapping Blocks. 510 db, MLCB
Deep Convolutional Nets as a Multi-Scale Matrix Mixture Factorization, Ankit B. Patel 511 c, Multiresolution
Poster session 511b, BayesOpt
Discussion, Q&A, etc. 512 cg, CognitiveComputation2
Ohad Shamir (Weizmann Inst.) Multiclass-Multilabel Classification with More Classes than Examples 511 f, Extreme
Poster session 1 511 a, MultiAgent
Poster highlights/Start of poster session 2 515 a, Neuroimaging2
Yoshua Bengio 514 bc, TransferAndMultitask
Accelerating exact MCMC with subsets of data by Ryan Adams (Harvard University) 513 ab, ScalableMC
Panel Discussion 515 bc, BayesNonparam
11:35am EST
11:40am EST
11:45am EST
11:50am EST
12:00pm EST
12:20pm EST
12:25pm EST
12:30pm EST
12:45pm EST
1:00pm EST
1:30pm EST
2:00pm EST
2:15pm EST
2:30pm EST
Durk Kingma 513 ef, BlackBox
Jeff Schneider 511b, BayesOpt
Asela Gunawardana (Microsoft) Evaluating Machine Learned User Experiences 511 f, Extreme
Invited Talk: Sarah Bird 511 d, MLSys
Fast Direct Methods for Gaussian Processes, Michael O'Neil 511 c, Multiresolution
Sanjeev Arora 513 cd, NonConvexOpt
Shai Ben-David 514 bc, TransferAndMultitask
Neural Machine Translation: Progress Report and Beyond, Kyunghyun Cho, NYU (Invited) 510 ac, RAM
Michael Bowling (University of Alberta) 511 a, MultiAgent
Daniel Lidar (TBA) 512 a, QuantumML
Dana Pe’er - TBD 510 db, MLCB
Invited Speaker: Wendy Cho 512 dh, Privacy
Poster Session 515 bc, BayesNonparam
Invited talk: Cheryl Corcoran 515 a, Neuroimaging2
2:40pm EST
2:52pm EST
3:00pm EST
Discussion 511 d, MLSys
Approximating Gaussian Processes with H^2 Matrices, Steffen Börm 511 c, Multiresolution
Marc Deisenroth 511b, BayesOpt
Noam Koenigstein (Microsoft) Implicit Feedback and Performance Evaluation in Recommender Systems 511 f, Extreme
Hossein Mobahi 513 cd, NonConvexOpt
Ruslan Salakhutdinov 514 bc, TransferAndMultitask
Lessons Learned from the PASCAL VOC Challenges, and Improving the Data Analytics Process, Chris Williams 512 e, CiML
Emily Fox 512 bf, NetworksInScience
David Woodruff, IBM Almaden 511 e, BigNeuro
Spotlights & Posters 514 a, Personalization
3:05pm EST
3:10pm EST
3:15pm EST
3:20pm EST
3:30pm EST
Contributed talk (Charles J. M. Mathy) 511 a, MultiAgent
Poster session 511b, BayesOpt
Sara Wade 515 bc, BayesNonparam
Van der Velde: Combinatorial structures and processing in Neural Blackboard Architectures 512 cg, CognitiveComputation2
Panel discussion with Asela, Noam, Armand Joulin and Patrice Simard chaired by Manik Varma 511 f, Extreme
Contributed - Sami Remes - Classification of weak multi-view signals by sharing factors in a mixture of Bayesian group factor analyzers 515 a, Neuroimaging2
Andrew Barron 513 cd, NonConvexOpt
Spotlight of Posters 514 bc, TransferAndMultitask
3:35pm EST
3:37pm EST
3:40pm EST
3:45pm EST
3:50pm EST
4:00pm EST
4:15pm EST
4:20pm EST
4:25pm EST
4:30pm EST
Matt Hoffman 511b, BayesOpt
Charles Elkan (Amazon/UCSD) Massive Sparse Multilabel Learning 511 f, Extreme
Invited Talk: Roxana Geambasu 511 d, MLSys
Francis Bach 511 c, Multiresolution
Chris Re 513 cd, NonConvexOpt
Ruslan Salakutdinov 513 ef, BlackBox
Frans Oliehoek (University of Amsterdam) 511 a, MultiAgent
Mohammad Amin (Quantum Boltzmann Machine) 512 a, QuantumML
Matthew Stephens - TBD 510 db, MLCB
Ambuj Tewari, Huitian Lei, & Susan Murphy. University of Michigan. 'From Ads to Interventions: Contextual Bandit Algorithms for Mobile Health'. (NIH application to 'Heartsteps') 514 a, Personalization
Invited Speaker: Max Ott 512 dh, Privacy
Richard G. Baraniuk, Rice University 511 e, BigNeuro
Tom Mitchell 515 a, Neuroimaging2
4:40pm EST
5:00pm EST
Discussion 511 d, MLSys
Amr Ahmed 515 bc, BayesNonparam
Academic Torrents: Scalable Data Distribution, Henry Z. Lo 512 e, CiML
Gary F. Marcus 512 cg, CognitiveComputation2
David Belanger (UMass) Scaling up Multilabel Classification using Structured Prediction Energy Networks 511 f, Extreme
Scaling Phenomena in Stochastic Topology, Sayan Mukherjee 511 c, Multiresolution
Kevin Chen 513 cd, NonConvexOpt
Percy Liang 514 bc, TransferAndMultitask
A Roadmap towards Machine Intelligence, Tomas Mikolov, Facebook AI Research (Invited) 510 ac, RAM
Deepak Agarwal 512 bf, NetworksInScience
Panel discussion 511b, BayesOpt
5:10pm EST
5:15pm EST
5:20pm EST
5:30pm EST
Discussion, Q&A, etc. 512 cg, CognitiveComputation2
Isabel Valera 515 bc, BayesNonparam
Michiel Stock (Ghent) A two-step method to incorporate task features for large output spaces 511 f, Extreme
Qiang Yang 514 bc, TransferAndMultitask
Open discussion 512 e, CiML
Panel discussion 2 + snacks and drinks + closing remarks 515 a, Neuroimaging2
Open problem session + closing remarks 513 cd, NonConvexOpt
Panel & Group Discussion on Conclusions & Future Directions. Finale Doshi-Velez, Ambuj Tewari, Joseph Jay Williams, Neil Heffernan 514 a, Personalization
5:35pm EST
5:40pm EST
5:45pm EST
5:50pm EST
5:55pm EST
6:00pm EST
6:15pm EST
6:20pm EST