Best Presentation Award

Based on a vote by workshop registrants, our Best Presentation Award goes to Tasneem Kaochar’s presentation, Challenges to Decoding the Intention Behind Natural Instruction.



Full schedule:

Day 1 (Saturday, 7/16)

8:30 - 9:00 : Registration

9:00 - 11:00: Session 1A

  1. •Introductory talk - W. Bradley Knox (30 min.) slides

  2. •Invited talk - Joanna Bryson (1 hr) slides

  3. •Towards Task Understanding through Multi-State Visuo-Spatial Perspective Taking for Human-Robot Interaction - Amit Kumar Pandey and Rachid Alami slides

11:00 - 11:30: Coffee break

11:30 - 13:00: Session 1B - Learning the meaning of human communication / Multiple teaching modalities (Session Chair - Joanna Bryson)

  1. •Game-based Language Tutoring - Katrien Beuls and Joris Bleys slides

  2. •Simultaneous Acquisition of Task and Feedback Models - Manuel Lopes, Thomas Cederbourg and Pierre-Yves Oudeyer slides

  3. •A Hidden Domain for Human and Electronic Students - Jeremy Ludwig, Alex Davis, Mark Abrams and Jon Curtis slides

13:00 - 14:30: Lunch break

14:30 - 16:30: Session 1C - Comparisons of teaching methods (Session Chair - Joanna Bryson)

  1. •Invited talk - Tom Dietterich (1 hr) slides

  2. •Assisting Machine Learning Through Shaping, Advice and Examples - Igor Karpov, Vinod Valsalam and Risto Miikkulainen slides

  3. •Understanding Human Teaching Modalities in Reinforcement Learning Environments: A Preliminary Report - W. Bradley Knox, Matthew Taylor and Peter Stone slides

16:30 - 17:00: Coffee break

17:00 - 18:30: Session 1D

  1. •Constraining the Size Growth of the Task Space with Socially Guided Intrinsic Motivation using Demonstrations - Sao Mai Nguyen, Adrien Baranes and Pierre-Yves Oudeyer slides

  2. •Invited short talk by Ian Fasel, who will then lead a discussion/panel on human behavioral research using learning agents (1 hr)


Day 2 (Sunday, 7/17)

8:30 - 9:00 : Registration

9:00 - 11:00: Session 2A - Reinforcement learning and learning from demonstration (Session Chair - Tom Dietterich)

  1. •Invited talk - Jan Peters (1 hr) slides

  2. •Learning Options through Human Interaction - Kaushik Subramanian, Charles L. Isbell and Andrea L. Thomaz slides

  3. •Batch, Off-policy and Model-Free Apprenticeship Learning - Edouard Klein, Matthieu Geist and Olivier Pietquin slides

11:00 - 11:30: Coffee break

11:30 - 13:00: Session 2B  - Learning from demonstration (Session Chair - Jan Peters)

  1. •Multiagent Supervised Training with Agent Hierarchies and Manual Behavior Decomposition - Keith Sullivan and Sean Luke slides

  2. •Towards a Unified Framework for Learning from Observation - Santiago Ontañón, Jose L. Montaña and Avelino Gonzalez slides

  3. •Robot Learning from Demonstration in the Force - Leonel Rozo, Pablo Jimenez and Carme Torras slides

13:00 - 15:00: Lunch break (extra long)

15:00 - 16:30: Session 2C - Natural Language

  1. •Invited talk - Dan Roth (1 hr) slides

  2. •Reciprocal Learning via Dialogue Interaction: Challenges and Prospects - Raquel Fernández, Staffan Larsson, Robin Cooper, Jonathan Ginzburg and David Schlangen slides

16:30 - 17:00: Coffee break

17:00 - 18:30: Session 2D - Natural Language

  1. •Challenges to Decoding the Intention Behind Natural Instruction - Raquel Torres Peralta, Tasneem Kaochar, Ian Fasel, Clay Morrison, Tom Walsh and Paul Cohen slides

  2. •Open discussion (1 hr)

19:00: Group dinner at Aribau-3 (the address is Carrer de Aribau 3)


Presentations of accepted papers will be 30 minutes.



Confirmed invited speakers:

  1. •Joanna Bryson (University of Bath)

  2. •Thomas G. Dietterich (Oregon State)

  3. •Ian Fasel (University of Arizona)

  4. •Jan Peters (Max-Planck Institute)

  5. •Dan Roth (University of Illinois - Urbana Champaign)



Voting for Best Presentation

Email your vote to Brenna Argall at brennadee.argall{at}epfl.ch between the end of the last paper presentation and midnight local time that night (Sunday, 7/17). Include paper title and authors as well as your full name. (Only paper presentations are eligible for the award. Authors cannot vote for themselves.) Only votes from workshop registrants will be counted. We will post the winner shortly thereafter.