Supporting materials for the paper
Proceeding of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021) - Adaptive Learning Agents (ALA) Workshop
Hossein Haeri, Reza Ahmadzadeh, Kshitij Jerath
University of Massachusetts Lowell
UPDATE (Jan 2024) : If your are interested in this work please check my thesis available here
To reference this work please cite the full version available on arXiv
This work has been presented at the AAMAS ALA 2021 workshop (presentation, slides)
a short 2-page version is available at the ACM digital library
Checkout the Github repo for codes and results:
https://github.com/hossein-haeri/reward-sharing-relational-networks
The proposed framework adds a reward sharing block to a typical MARL flow diagram which takes a vector of individual rewards and a user-defined relational network as inputs and gives a vector of shared rewards to the policy optimizer.
Each of these relational networks is used as an RSRN in the 3-agent MARL system, and the resulting emergent behaviors are examined. An agent’s actions may be governed by the rewards obtained by more than one agent (including itself). An arrow from an agent to itself indicates its actions would maximize the rewards it obtains for itself. An arrow directed from a first agent to a second represents that the first agent’s actions are governed by the rewards obtained by the second agent, i.e. an agent it ‘cares’ about.
RSRN_Demo.mp43 agents try to cover 3 landmarks while agent 3 (yellow) is systematically slowed down by limiting its possible actions.