A3C-S: Automated Agent Accelerator Co-Search towards Efficient Deep Reinforcement Learning

TL;DR

This paper introduces A3C-S, an automated framework that co-searches for optimal deep reinforcement learning agents and hardware accelerators, significantly improving performance and efficiency for resource-constrained devices.

Contribution

A3C-S is the first framework to automatically co-search DRL agents and accelerators, optimizing both test scores and hardware efficiency simultaneously.

Findings

A3C-S outperforms existing methods in test scores.

A3C-S achieves higher hardware efficiency.

Experimental results validate the framework's superiority.

Abstract

Driven by the explosive interest in applying deep reinforcement learning (DRL) agents to numerous real-time control and decision-making applications, there has been a growing demand to deploy DRL agents to empower daily-life intelligent devices, while the prohibitive complexity of DRL stands at odds with limited on-device resources. In this work, we propose an Automated Agent Accelerator Co-Search (A3C-S) framework, which to our best knowledge is the first to automatically co-search the optimally matched DRL agents and accelerators that maximize both test scores and hardware efficiency. Extensive experiments consistently validate the superiority of our A3C-S over state-of-the-art techniques.