Safe adaptation in multiagent competition

03/14/2022
by   Macheng Shen, et al.
0

Achieving the capability of adapting to ever-changing environments is a critical step towards building fully autonomous robots that operate safely in complicated scenarios. In multiagent competitive scenarios, agents may have to adapt to new opponents with previously unseen behaviors by learning from the interaction experiences between the ego-agent and the opponent. However, this adaptation is susceptible to opponent exploitation. As the ego-agent updates its own behavior to exploit the opponent, its own behavior could become more exploitable as a result of overfitting to this specific opponent's behavior. To overcome this difficulty, we developed a safe adaptation approach in which the ego-agent is trained against a regularized opponent model, which effectively avoids overfitting and consequently improves the robustness of the ego-agent's policy. We evaluated our approach in the Mujoco domain with two competing agents. The experiment results suggest that our approach effectively achieves both adaptation to the specific opponent that the ego-agent is interacting with and maintaining low exploitability to other possible opponent exploitation.

READ FULL TEXT

page 4

page 5

research
11/12/2020

Learning Latent Representations to Influence Multi-Agent Interaction

Seamlessly interacting with humans or robots is hard because these agent...
research
06/15/2023

Behavioral Cloning via Search in Embedded Demonstration Dataset

Behavioural cloning uses a dataset of demonstrations to learn a behaviou...
research
08/26/2020

Safe Model-Based Meta-Reinforcement Learning: A Sequential Exploration-Exploitation Framework

Safe deployment of autonomous robots in diverse environments requires ag...
research
12/07/2021

MESA: Offline Meta-RL for Safe Adaptation and Fault Tolerance

Safe exploration is critical for using reinforcement learning (RL) in ri...
research
03/24/2022

Differential Assessment of Black-Box AI Agents

Much of the research on learning symbolic models of AI agents focuses on...
research
12/13/2017

Adaptation to criticality through organizational invariance in embodied agents

Many biological and cognitive systems do not operate deep within one or ...
research
06/17/2022

Beyond Rewards: a Hierarchical Perspective on Offline Multiagent Behavioral Analysis

Each year, expert-level performance is attained in increasingly-complex ...

Please sign up or login with your details

Forgot password? Click here to reset