UAI 2017 Program Schedule


Note: This is a preliminary outline of the schedule and is likely to undergo slight modifications.

August 11th

Time Event
8:45 - 10:15 Tutorial 1
10:35 - 12:05 Tutorial 2
14:05 - 15:35 Tutorial 3
16:00 - 17:30 Tutorial 4

August 12th

Time Event
07:30 - 08:30 Opening hours registration desk
08:30 - 08:40 Welcome
08:40 - 09:40 Keynote talk
09:40 - 10:40 Oral Session
10:40 - 11:10 Coffee Break
11:10 - 12:10 Oral Session
12:10 - 14:00 Lunch break
14:00 - 15:00 Keynote talk
15:00 - 16:00 Oral Session
16:00 - 16:20 Coffee Break
16:20 - 17:20 Oral Session
17:20 - 17:50 Poster Spotlights
17:50 - 19:50 Poster Session

August 13th

Time Event
08:30 - 09:30 Keynote talk
09:30 - 10:30 Oral Session
10:30 - 11:00 Coffee Break
11:00 - 12:20 Oral Session
12:20 - 14:10 Lunch break
14:10 - 15:10 Keynote talk
15:10 - 15:40 Poster Spotlights
15:40 - 16:00 Coffee Break
16:00 - 18:00 Poster Session
19:00 Banquet Boarding

August 14th

Time Event
08:30 - 09:30 Keynote talk
09:30 - 10:30 Oral Session
10:30 - 11:00 Coffee Break
11:00 - 12:20 Oral Session
12:20 - 14:10 Lunch break
14:10 - 15:10 Oral Session
15:10 - 15:40 Poster spotlights
15:40 - 16:00 Coffee Break
16:00 - 16:45 Business meeting
16:00 - 18:00 Poster Session

Detailed Program Schedule


August 11th

Time Event
8:45 - 10:15 Tutorial 1
  • John C. Duchi: Methods and models for large-scale optimization
10:35 - 12:05 Tutorial 2
  • Arthur Gretton: Representing and comparing probabilities with (and without) kernels
14:05 - 15:35 Tutorial 3
  • Shakir Mohamed and Danilo Rezende: Deep Generative Models
16:00 - 17:30 Tutorial 4
  • Suchi Saria: Machine learning in healthcare

August 12th

Time Event
07:30 - 08:30 Opening hours registration desk
08:30 - 08:40 Welcome
08:40 - 09:40 Keynote talk
  • Prof. Leslie Pack Kaelbling: Intelligent Robots in an Uncertain World
09:40 - 10:40 Oral Session Deep Models
  1. Inverse reinforcement learning via deep gaussian process
  2. Holographic feature representations of deep networks
  3. Computing nonvacuous generalization bounds for deep stochastic neural networks with many more parameters than training data
10:40 - 11:10 Coffee Break
11:10 - 12:10 Oral Session Machine Learning
  1. Provable inductive robust PCA via iterative hard thresholding
  2. Near orthogonality regularization in kernel methods
  3. How good are my predictions efficiently approximating precision recall curves for massive datasets
12:10 - 14:00 Lunch break
14:00 - 15:00 Keynote talk
  • Prof. Amir Globerson: TBA
15:00 - 16:00 Oral Session Inference
  1. On loopy belief propagation local stability analysis for non vanishing fields
  2. Improving optimization based approximate inference by clamping variables
  3. Approximation complexity of maximum a posteriori inference in sum product networks
16:00 - 16:20 Coffee Break
16:20 - 17:20 Oral Session Learning
  1. Learning the structure of probabilistic sentential decision diagrams
  2. A probabilistic framework for multilabel learning with unseen labels
  3. Hybrid deep discriminative generative models for semi supervised learning
17:20 - 17:50 Poster Spotlights
17:50 - 19:50 Poster Session

August 13th

Time Event
08:30 - 09:30 Keynote talk
  • Prof. Christopher Re: Snorkel: Beyond Hand-labeled Data
09:30 - 10:30 Oral Session Representations
  1. Why rules are complex real valued probabilistic logic programs are not fully expressive
  2. Interpreting lion behaviour as probabilistic programs
  3. Decoupling homophily and reciprocity with latent space network models
10:30 - 11:00 Coffee Break
11:00 - 12:20 Oral Session Reinforcement Learning
  1. Online constrained model based reinforcement learning
  2. A reinforcement learning approach to weaning of mechanical ventilation in intensive care units
  3. Near optimal interdiction of factored MDPs
  4. Importance sampling for fair policy selection
12:20 - 14:10 Lunch break
14:10 - 15:10 Keynote talk
  • Prof. Katherine Heller: TBA
15:10 - 15:40 Poster Spotlights
15:40 - 16:00 Coffee Break
16:00 - 18:00 Poster Session
19:00 Banquet Boarding
  • Prof. Terry Speed: 15 minutes on artificial intelligence and statistical models

August 14th

Time Event
08:30 - 09:30 Keynote talk
  • Prof. Terry Speed: Two current analysis challenges: Single Cell Omics and Nanopore Long-read Sequence Data
09:30 - 10:30 Oral Session Causality
  1. Learning treatment response models from multivariate longitudinal data
  2. Interpreting and using CPDAGs with background knowledge
  3. Causal consistency of structural equation models
10:30 - 11:00 Coffee Break
11:00 - 12:20 Oral Session Sampling
  1. Stein variational adaptive importance sampling
  2. Continuously tempered Hamiltonian Monte Carlo
  3. Balanced minibatch sampling for SGD using determinantal point processes
  4. An efficient minibatch acceptance test for Metropolis-Hastings
12:20 - 14:10 Lunch break
14:10 - 15:10 Oral Session Bandits
  1. Stochastic bandit models for delayed conversions
  2. A practical method for solving contextual bandit problems using decision trees
  3. Analysis of Thompson sampling for stochastic sleeping bandits
15:10 - 15:40 Poster spotlights
15:40 - 16:00 Coffee Break
16:00 - 16:45 Business meeting
16:00 - 18:00 Poster Session

Poster Sessions August 12th

  1. Regret minimization algorithms for the followers behaviour identification in leadership games
  2. On the complexity of nash equilibrium reoptimization
  3. Shortest path under uncertainty exploration versus exploitation
  4. Learning with confident examples rank pruning for robust classification with noisy labels
  5. Montecarlo tree search using batch value of perfect information
  6. Submodular variational inference for network reconstruction
  7. Bayesian inference of log determinants
  8. Fast amortized inference and learning in loglinear models with randomly perturbed nearest neighbor search
  9. Supervised restricted boltzmann machines
  10. Safe semisupervised learning of sumproduct networks
  11. Green generative modeling recycling dirty data using recurrent variational autoencoders
  12. Approximate evidential reasoning using local conditioning and conditional belief functions
  13. Differentially private variational inference for nonconjugate models
  14. Value directed exploration in multiarmed bandits with structured priors
  15. Learning approximately objective priors
  16. Learning to draw samples with amortized stein variational gradient descent
  17. A tractable probabilistic model for subset selection
  18. Structure learning of linear gaussian structural equation models with weak edges
  19. Satbased causal discovery under weaker assumptions
  20. Learning to acquire information

Poster Sessions August 13th

  1. Frosh: faster online sketching hashing
  2. Self-discrepancy conditional independence test
  3. Towards conditional independence test for relational data
  4. Autogp: exploring the capabilities and limitations of gaussian process models
  5. A fast algorithm for matrix eigendecomposition
  6. Branch and bound for regular bayesian network structure learing
  7. Effective sketching methods for value function approximation
  8. Stochastic lbfgs revisited improved convergence rates and practical acceleration strategies
  9. The binomial block bootstrap estimator for evaluating loss on dependent clusters
  10. Datadependent sparsity for subspace clustering
  11. Weighted model counting with function symbols
  12. Triply stochastic gradients on multiple kernel learning
  13. Coupling adaptive batch sizes with learning rates
  14. Composing inference algorithms as program transformations
  15. Iterative decomposition guided variable neighborhood search for graphical model energy minimization
  16. Fair optimal stopping policy for matching with mediator
  17. Exact inference for relational graphical models with interpreted functions lifted probabilistic inference modulo theories
  18. Neighborhood regularized ellgraph
  19. Feature-to-feature regression for a twostep conditional independence test
  20. Algebraic equivalence class selection for linear structural equation models

Poster Sessions August 14th

  1. The total belief theorem
  2. Complexity of solving decision trees with skewsymmetric bilinear utility
  3. Stochastic segmentation trees for multiple ground truths
  4. Efficient online learning for optimizing value of information theory and application to interactive troubleshooting
  5. Counting markov equivalence classes by number of immoralities
  6. Realtime resource allocation for tracking systems
  7. Synthesis of strategies in influence diagrams
  8. Embedding senses via dictionary bootstrapping
  9. Importance sampled stochastic optimization for variational inference
  10. Multi-dueling bandits with dependent arms
  11. Convex-constrained sparse additive modeling and its extensions
  12. Stein variational policy gradient
  13. Causal discovery from temporally aggregated time series
  14. Efficient solutions for stochastic shortest path problems with dead ends
  15. Probabilistic program abstractions
  16. Communication-efficient distributed primaldual algorithm for saddle point problem
  17. Robust model equivalence using stochastic bisimulation for nagent interactive dids
  18. Adversarial sets for regularising neural link predictors




Golden Sponsor



Golden Sponsor



Golden Sponsor



Bronze Sponsor



Startup Sponsor



Media Sponsor