Workshop
Learning by Instruction
Shashank Srivastava · Igor Labutov · Bishan Yang · Amos Azaria · Tom Mitchell

Sat Dec 8th 08:00 AM -- 06:30 PM @ Room 516 AB
Event URL: https://sites.google.com/view/lbi2018/ »

Today machine learning is largely about pattern discovery and function approximation. But as computing devices that interact with us in natural language become ubiquitous (e.g., Siri, Alexa, Google Now), and as computer perceptual abilities become more accurate, they open an exciting possibility of enabling end-users to teach machines similar to the way in which humans teach one another. Natural language conversation, gesturing, demonstrating, teleoperating and other modes of communication offer a new paradigm for machine learning through instruction from humans. This builds on several existing machine learning paradigms (e.g., active learning, supervised learning, reinforcement learning), but also brings a new set of advantages and research challenges that lie at the intersection of several fields including machine learning, natural language understanding, computer perception, and HCI.

The aim of this workshop is to engage researchers from these diverse fields to explore fundamental research questions in this new area, such as:
How do people interact with machines when teaching them new learning tasks and knowledge?
What novel machine learning models and algorithms are needed to learn from human instruction?
What are the practical considerations towards building practical systems that can learn from instruction?

08:30 AM Introduction (Welcome)
08:35 AM Teaching Machines like we Teach People (Talk from Organizers)
09:00 AM Mapping Navigation Instructions to Continuous Control (Invited Talk) Yoav Artzi
09:30 AM An Cognitive Architecture Approach to Interactive Task Learning (Invited Talk) John Laird
10:00 AM Compositional Imitation Learning: Explaining and executing one task at a time (Contributed Talk) Thomas Kipf
10:15 AM Learning to Learn from Imperfect Demonstrations (Contributed Talk) Ge Yang, Chelsea Finn
11:00 AM Natural Language Supervision (Invited Talk) Percy Liang
11:30 AM Control Algorithms for Imitation Learning from Observation (Invited Talk) Peter Stone
12:00 PM From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following (Contributed talk) Justin Fu
12:15 PM Teaching Multiple Tasks to an RL Agent using LTL (Contributed Talk) Rodrigo Toro Icarte, Sheila McIlraith
01:30 PM Meta-Learning to Follow Instructions, Examples, and Demonstrations (Invited Talk) Sergey Levine
02:00 PM Learning to Understand Natural Language Instructions through Human-Robot Dialog (Invited Talk) Ray Mooney
02:30 PM The Implicit Preference Information in an Initial State (Contributed Talk) Rohin Shah
02:45 PM Modelling User's Theory of AI's Mind in Interactive Intelligent Systems (Contributed Talk) Tomi Peltola
03:30 PM Poster Session <span> <a href="#"></a> </span>
Carl Trimbach, Mennatullah Siam, Rodrigo Toro Icarte, Falcon Dai, Sheila McIlraith, Matthew Rahtz, Rob Sheline, Chris MacLellan, Carolin Lawrence, Stefan Riezler, Dylan Hadfield-Menell, Fang-I Hsiao
04:15 PM Assisted Inverse Reinforcement Learning (Contributed Talk) Adish Singla, Rati Devidze
04:30 PM Teaching through Dialogue and Games (Invited Talk) Jason E Weston
05:00 PM Panel Discussion (Discussion Panel)

Author Information

Shashank Srivastava (Microsoft Research)
Igor Labutov (Cornell University)
Bishan Yang (Cornell University)
Amos Azaria (Ariel University)
Tom Mitchell (Carnegie Mellon University)

More from the Same Authors