ALT 2014 Program

• Tuesday 7th October 2014, 18:00 - 20:00: Registration Desk
• Tuesday 7th October 2014, 19:00 - 21:00: Welcome Reception
• Wednesday 8th October 2014, 8:00 - 18:00: Registration Desk
• Wednesday 8th October 2014, 9:00 - 10:30: Tutorial ALT
  Eyke Hüllermeier: A Survey of Preference-based Online Learning with Bandit Algorithms
  
  
• Wednesday 8th October 2014, 10:30 - 11:00: Coffee Break
  
• Wednesday 8th October 2014,11:00 - 12:30: Tutorial DS
  Anuška Ferligoj: Social Network Analysis
• Wednesday 8th October 2014, 12:30 - 14:00: Lunch
  
• Wednesday 8th October 2014, 14:00 - 15:00:
  Invited Talk ALT/DS Zoubin Ghahramani:
  Building an Automated Statistician
  
• Wednesday 8th October 2014, 15:00 - 15:30: Coffee Break
  
• Wednesday 8th October 2014, 15:30 - 17:00:
  ALT/DS Session 1 – Inductive Inference
  
  • Timo Kötzing and Raphaela Palenta:
    A Map of Update Constraints in Inductive Inference
  • Sanjay Jain, Timo Kötzing, Junqi Ma, and Frank Stephan:
    On the Role of Update Constraints and Text-Types in Iterative Learning
  • Sanjay Jain and Efim Kinber:
    Parallel Learning of Automatic Classes of Languages
• Wednesday 8th October 2014, 17:00 - 18:00: Steering Committee Meeting
• Wednesday 8th October 2014, 19:00 - 21:00: Poster Session with Cocktail Party
  
  
• Thursday, 9th October, 9:00 - 10:00:
  Invited Talk Luc Devroye:
  Cellular Tree Classifiers
  
• Thursday, 9th October, 10:00 - 10:30:
  E.M. Gold Award Talk (ALT)
  Hasan Abasi, Ali Abdi and Nader Bshouty:
  Learning Boolean Halfspaces with Small Weights from Membership Queries
  
• Thursday, 9th October, 10:30 - 11:00: Coffee Break
  
• Thursday, 9th October, 11:00 - 12:30:
  ALT/DS Session 2 – Online Learning
  • Tor Lattimore, Andras Gyorgy and Csaba Szepesváriri:
    On Learning the Optimal Waiting Time
  • Nir Ailon, Kohei Hatano and Eiji Takimoto:
    Bandit Online Optimization Over the Permutahedron
  • Marcus Hutter:
    Offline to Online Conversion
• Thursday, 9th October, 12:30 - 14:00: Lunch
  
• Thursday, 9th October, 14:00 - 15:30:
  ALT/DS Session 3 – Privacy, Reinforcement Learning
  • Christos Dimitrakakis, Blaine Nelson, Aikaterini Mitrokotsa and Benjamin Rubinstein:
    Robust and Private Bayesian Inference
  • Ronald Ortner, Odalric-Ambrym Maillard and Daniil Ryabko:
    Selecting Near-Optimal Approximate State Representations in Reinforcement Learning
  • Prashanth L.A.:
    Policy Gradients for CVaR-Constrained MDPs
• Thursday, 9th October, 15:30 - 16:00: Coffee Break
• Thursday, 9th October, 16:00 - 17:00:
  ALT/DS Session 4 – General RL
  • Tor Lattimore and Marcus Hutter:
    Bayesian Reinforcement Learning with Exploration
  • Marcus Hutter:
    Extreme State Aggregation beyond MDPs
• Thursday, 9th October, 17:00 - 17:30: Business Meeting
• Thursday, 9th October, 18:00: Departure to Ljubljana
• Thursday, 9th October, 20:00 - 22:00: Conference Dinner (Ljubljana castle)
  
  
• Friday, 10th October, 9:00 - 10:00:
  Invited Talk Michel Dumontier:
  Semantic Approaches for Biomedical Knowledge Discovery
• Friday, 10th October, 10:00 - 10:30:
  Carl H. Smith Award Talk (DS)
  Pedro Pereira, Rita Ribeiro, and Joao Gama:
  Failure Prediction - an Application in the Railway Industry
  
• Friday, 10th October, 10:30 - 11:00: Coffee Break
  
• Friday, 10th October, 11:00 - 12:30:
  ALT/DS Session 5 – Statistical Learning Theory
  • Andreas Maurer:
    A Chain Rule for the Expected Suprema of Gaussian Processes
  • Vitaly Kuznetsov and Mehryar Mohri:
    Generalization Bounds for Time Series Prediction with Non-stationary Processes
  • Rahim Samei, Boting Yang and Sandra Zilles:
    Generalizing Labeled and Unlabeled Sample Compression to Multi-label Concept Classes
• Friday, 10th October, 12:30 - 14:00: Lunch
  
• Friday, 10th October, 14:00 - 15:30:
  ALT/DS Session 6 – MDL, Kolmogorov Complexity
  • Jan Leike and Marcus Hutter:
    Indefinitely Oscillating Martingales
  • Peter Bloem, Francisco Mota, Steven De Rooij, Luis Antunes and Pieter Adriaans:
    A Safe Approximation for Kolmogorov Complexity
  • Laurent Bienvenu, Benoît Monin and Alexander Shen:
    Algorithmic Identification of Probabilities is Hard
• Friday, 10th October, 15:30 - 16:00: Coffee Break
  
• Friday, 10th October, 16:00 - 17:30:
  ALT/DS Session 7 – Clustering, Exact Learning from Queries
  • Behnam Neyshabur, Yury Makarychev and Nathan Srebro:
    Clustering, Hamming Embedding, Generalized LSH and the Max Norm
  • Hasan Abasi, Nader Bshouty and Hanna Mazzawi:
    On Exact Learning Monotone DNF from Membership Queries
  • Dana Angluin and Dana Fisman:
    Learning Regular Omega Languages