[agents] Challenges and Opportunities in Multiagent RL: Branislav Bosansky

[Frans Oliehoek]

Dear all,


Happy to announce the next speaker in our virtual seminars on the 
Challenges and Opportunities for Multiagent Reinforcement Learning (COMARL):


Speaker: Branislav Bosansky

Title: Solving Dynamic Games with Imperfect Information

When: November 18th


Abstract:

Finding optimal strategies for dynamic multi-agent interactions, where 
agents have only partial observations about the environment, is one of 
today's challenges. Even the cases with two agents and strictly 
competitive interactions (i.e., zero-sum games) are difficult -- 
especially if we consider interactions that either require many turns to 
complete or do not have a limited number of turns. At the same time, we 
want algorithms with bounded error to know how close to (or far away 
from) the optimum the found strategies are.


  From the game-theoretic perspective, we can model two-agent strictly 
competitive interactions as zero-sum partially observable stochastic 
games (zs-POSGs) with the infinite or indefinite horizon. Since even 
zs-POSGs can be undecidable, we pose further restrictions that allow us 
to design and implement search algorithms that are guaranteed to 
converge to optimal strategies and have a bounded error. Our algorithms 
are inspired by Heuristic Search Value Iteration (HSVI) for 
partially-observable Markov decision processes (POMDPs), however, 
significantly modified to solve games where (a) only one player has 
partial information or (b) where both players have partial information 
but all observations are public.


In the talk, I will describe the key characteristics and the schema of 
all of our algorithms and identify future directions to scale up and/or 
generalize our algorithms.


Date & time: November 18th 10AM EDT / 3PM UTC / 4PM CET

How to 
attend:https://sites.google.com/view/comarl-seminars/how-to-attend 
<https://sites.google.com/view/comarl-seminars/how-to-attend>


Calendar with talks and Google Meet 
links:https://calendar.google.com/calendar/u/0?cid=Y19uZm9xdHZvZWw3Z3R0NDg0aGduZDUxc3U1NEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t 
<https://calendar.google.com/calendar/u/0?cid=Y19uZm9xdHZvZWw3Z3R0NDg0aGduZDUxc3U1NEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t>


We look forward to seeing you there!


Best regards from the organizers,


Chris Amato (Northeastern University),

Marta Garnelo (DeepMind),

Robert Loftin (TU Delft),

Frans Oliehoek (TU Delft),

Shayegan Omidshafiei (Google),

Karl Tuyls (DeepMind)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cs.umbc.edu/pipermail/agents/attachments/20211111/43170b3f/attachment.html>