Neuromorphic chip runs AI learning tasks faster and with much less power
All On-Board: Fully On-Chip Neuromorphic Q-Learning with Embedded CartPole Simulation
Neural and Evolutionary ComputingMachine Learning
Summary
As AI models use more energy, scientists want to find ways to make them use less power. The authors designed a brain-inspired computer chip that can learn to balance a pole in a classic test called CartPole. Their chip learned the task just as well as a regular computer but used 100 times less power and trained twice as fast. This shows such special chips could help build energy-efficient AI devices that work inside small gadgets or robots.
What this means in practice
- •For embedded systems developers: Design low-power AI controllers that run real-time reinforcement learning entirely on a single chip for devices with limited energy and compute resources.
- •For robotics engineers: Implement fast and energy-efficient neuromorphic learning agents that can train and adapt control policies onboard robots without external computation.
Authors
Steven C. Nesbit, Giovanni T. Michel, Gerd J. Kunde, Edward Kim, Andrew T. Sornborger
Abstract
As AI models grow in size and usage, their energy demands increase dramatically, raising sustainability and economic concerns. Neuromorphic hardware, inspired by the energy efficiency of the brain, seeks to address this challenge by offering low-power, fast-processing alternatives to conventional computing. Such hardware is particularly well-suited to control systems deployed in resource-constrained environments, which are best trained via reinforcement learning (RL). This contribution presents the design and implementation of a fully on-chip, closed-loop Loihi 2 RL agent. Our neuromorphic circuit consists of a fully embedded Q-learning algorithm and an on-chip simulation of the CartPole-v0 environment on Loihi 2. Our Q-learning algorithm trained the same number of successful agents as the CPU implementation in only half the execution time and with two orders of magnitude less dynamic power. These findings demonstrate the viability of RL on neuromorphic hardware and highlight its promise for building energy-efficient, real-time, embedded AI systems.