EARL: Energy-Aware Optimization of Liquid State Machines for Pervasive AI

TL;DR

This paper introduces EARL, an energy-aware reinforcement learning framework that optimizes Liquid State Machines for low-power, resource-constrained on-device AI, significantly improving accuracy and reducing energy use and optimization time.

Contribution

The paper presents a novel energy-aware optimization framework combining Bayesian optimization and reinforcement learning for Liquid State Machines, addressing hyperparameter sensitivity and energy constraints.

Findings

Achieves 6-15% higher accuracy

Reduces energy consumption by 60-80%

Cuts optimization time by up to an order of magnitude

Abstract

Pervasive AI increasingly depends on on-device learning systems that deliver low-latency and energy-efficient computation under strict resource constraints. Liquid State Machines (LSMs) offer a promising approach for low-power temporal processing in pervasive and neuromorphic systems, but their deployment remains challenging due to high hyperparameter sensitivity and the computational cost of traditional optimization methods that ignore energy constraints. This work presents EARL, an energy-aware reinforcement learning framework that integrates Bayesian optimization with an adaptive reinforcement learning based selection policy to jointly optimize accuracy and energy consumption. EARL employs surrogate modeling for global exploration, reinforcement learning for dynamic candidate prioritization, and an early termination mechanism to eliminate redundant evaluations, substantially reducing computational overhead. Experiments on three benchmark datasets demonstrate that EARL achieves 6 to 15 percent higher accuracy, 60 to 80 percent lower energy consumption, and up to an order of magnitude reduction in optimization time compared to leading hyperparameter tuning frameworks. These results highlight the effectiveness of energy-aware adaptive search in improving the efficiency and scalability of LSMs for resource-constrained on-device AI applications.