An FPGA-Based On-Device Reinforcement Learning Approach using Online Sequential Learning

TL;DR

This paper presents a lightweight FPGA-based reinforcement learning method using OS-ELM that is more resource-efficient and faster than traditional DQN approaches, suitable for low-cost edge devices.

Contribution

It introduces a novel on-device reinforcement learning approach leveraging OS-ELM and regularization techniques, optimized for FPGA implementation on low-cost hardware.

Findings

Achieves 29.77x faster training on CartPole-v0 compared to DQN.

Demonstrates stable learning with spectral normalization and L2 regularization.

Successfully implements reinforcement learning on PYNQ-Z1 FPGA platform.

Abstract

DQN (Deep Q-Network) is a method to perform Q-learning for reinforcement learning using deep neural networks. DQNs require a large buffer and batch processing for an experience replay and rely on a backpropagation based iterative optimization, making them difficult to be implemented on resource-limited edge devices. In this paper, we propose a lightweight on-device reinforcement learning approach for low-cost FPGA devices. It exploits a recently proposed neural-network based on-device learning approach that does not rely on the backpropagation method but uses OS-ELM (Online Sequential Extreme Learning Machine) based training algorithm. In addition, we propose a combination of L2 regularization and spectral normalization for the on-device reinforcement learning so that output values of the neural network can be fit into a certain range and the reinforcement learning becomes stable. The proposed reinforcement learning approach is designed for PYNQ-Z1 board as a low-cost FPGA platform. The evaluation results using OpenAI Gym demonstrate that the proposed algorithm and its FPGA implementation complete a CartPole-v0 task 29.77x and 89.40x faster than a conventional DQN-based approach when the number of hidden-layer nodes is 64.