ALT 2015 Programme

Saturday 3 October 2015

17:00 - 19:00: Reception

Sunday 4 October 2015

7:00 - 8:30: Breakfast

8:30 - 8:45: Opening

8:45 - 9:45: Keynote 1: Sham M. Kakade (University of Washington): Finding Hidden Structure in Data with Tensor Decompositions.

9:45 - 10:35: Session 1: Inductive Inference

• Priced Learning
  Sanjay Jain, Junqi Ma and Frank Stephan
• Combining Models of Approximation with Partial Learning
  Ziyuan Gao, Frank Stephan and Sandra Zilles

10:35 - 10:55: Coffee Break

10:55 - 12:35: Session 2: Online Learning and Stochastic Optimization

• Scale-Free Algorithms for Online Linear Optimization
  Francesco Orabona and David Pal
• Online Learning in Markov Decision Processes with Continuous Actions
  Yi-Te Hong and Chi-Jen Lu
• Adaptive Sampling for Incremental Optimization using Stochastic Gradient Descent
  Guillaume Papa, Pascal Bianchi and Stephan Clemençon
• Online Linear Optimization for Job Scheduling under Precedence Constraints
  Takahiro Fujita, Kohei Hatano, Shuji Kijima and Eiji Takimoto

12:35 - 14:00: Lunch Break

14:00 - 15:15: Session 3: Statistical Learning Theory 1

• Learning with a Drifting Target Concept
  Steve Hanneke, Varun Kanade and Liu Yang
• Learning with Deep Cascades
  Giulia Desalvo, Mehryar Mohri and Umar Syed
• Bounds on the Minimax Rate for Estimating a Prior over a VC Class from Independent Learning Tasks
  Liu Yang, Steve Hanneke and Jaime Carbonell

15:15 - 15:35: Coffee Break

15:35 - 16:35: Business Meeting

16:35 - 17:30: Reception Hosted by ISM Canada

Monday 5 October 2015

7:00 - 8:45: Breakfast

8:45 - 9:45: Keynote 2: Kiri L. Wagstaff (California Institute of Technology): Overcoming Obstacles to the Adoption of Machine Learning by Domain Experts.

9:45 - 10:35: Session 4: Learning from Queries and Teaching Complexity

• Nonadaptive Learning of a Hidden Hypergraph
  Nader Bshouty, Hasan Abasi and Hanna Mazzawi
• On the Teaching Complexity of Linear Sets
  Ziyuan Gao, Hans Simon and Sandra Zilles

10:35 - 10:55: Coffee Break

10:55 - 11:25: Carl Smith Award Winner's Presentation

11:25 - 12:40: Session 5: Statistical Learning Theory 2

• On the Rademacher Complexity of Weighted Automata
  Borja Balle and Mehryar Mohri
• Multi-task and Lifelong Learning of Kernels
  Anastasia Pentina and Shai Ben-David
• Permutational Rademacher Complexity: a New Complexity Measure for Transductive Learning
  Ilya Tolstikhin, Nikita Zhivotovskiy and Gilles Blanchard

12:40 - 14:00: Lunch

14:00 - 15:00: Keynote 3: Cynthia Rudin (Massachusetts Intitute of Technology): Turning Prediction Tools Into Decision Tools.

15:00 - 15:30: EM Gold Award Winner's Presentation:

• Exact Learning Multivalued Dependencies
  Montserrat Hermo and Ana Ozaki

15:30 - 15:50: Coffee Break

15:50 - 17:05: Session 6: Computational Learning Theory

• Learning a Random DFA from Uniform Strings and State Information
  Dana Angluin and Dongqu Chen
• Hierarchical Design of Fast Minimum Disagreement Algorithms
  Malte Darnstädt, Christoph Ries and Hans Simon
• Interactive Clustering of Linear Classes and Cryptographic Lower Bounds
  Ádám D. Lelkes and Lev Reyzin

17:05 - 18:00: Late Breaking Papers

18:30 - 21:30: Banquet

Tuesday 6 October 2015

7:00 - 8:45: Breakfast

8:45 - 9:45: Keynote 4: Inderjit Dhillon (University of Texas, Austin): Bilinear Prediction using Low Rank Models.

9:45 - 10:35: Session 7: Dimensionality Reduction

• Subsampling in Smoothed Range Spaces
  Jeff Phillips and Yan Zheng
• Information Preserving Dimensionality Reduction
  Shrinu Kushagra and Shai Ben-David

10:35 - 10:55: Coffee Break

10:55 - 12:20: Session 8: Kolmogorov Complexity and Algorithmic Information Theory

• Solomonoff Induction Violates Nicod's Criterion
  Jan Leike and Marcus Hutter
• On the Computability of Solomonoff Induction and Knowledge-Seeking
  Jan Leike and Marcus Hutter
• Two Problems for Sophistication
  Peter Bloem and Steven De Rooij