Safety-Aware Multi-Agent Apprenticeship Learning

TL;DR

This paper extends safety-aware apprenticeship learning from single-agent to multi-agent systems, enhancing reward extraction and learning efficiency through novel frameworks and empirical evaluation.

Contribution

It introduces a multi-agent extension to inverse reinforcement learning for safety, along with a new learning framework and empirical performance assessment.

Findings

Successful extension of reward functions to multi-agent scenarios

Improved learning efficiency demonstrated empirically

Framework applicable to safety-critical multi-agent systems

Abstract

Our objective of this project is to make the extension based on the technique mentioned in the paper "Safety-Aware Apprenticeship Learning" to improve the utility and the efficiency of the existing Reinforcement Learning model from a Single-Agent Learning framework to a Multi-Agent Learning framework. Our contributions to the project are presented in the following bullet points: 1. Regarding the fact that we will add an extension to the Inverse Reinforcement Learning model from a Single-Agent scenario to a Multi-Agentscenario. Our first contribution to this project is considering the case of extracting safe reward functions from expert behaviors in a Multi-Agent scenario instead of being from the Single-Agent scenario. 2. Our second contribution is extending the Single-Agent Learning Framework to a Multi-Agent Learning framework and designing a novel Learning Framework based on the extension in the end. 3. Our final contribution to this project is evaluating empirically the performance of my extension to the Single-Agent Inverse Reinforcement Learning framework.