Brain-Inspired Hardware for Artificial Intelligence: Accelerated Learning in a Physical-Model Spiking Neural Network
This work addresses the need for low-power, adaptive computing substrates for AI, though it is incremental as it builds on existing neuromorphic approaches.
The researchers tackled the challenge of creating efficient brain-inspired hardware for AI by developing a neuromorphic chip that accelerates spiking neural network dynamics 1000-fold relative to biological time, enabling it to learn to play Pong through reinforcement learning with high energy efficiency and noise resilience.
Future developments in artificial intelligence will profit from the existence of novel, non-traditional substrates for brain-inspired computing. Neuromorphic computers aim to provide such a substrate that reproduces the brain's capabilities in terms of adaptive, low-power information processing. We present results from a prototype chip of the BrainScaleS-2 mixed-signal neuromorphic system that adopts a physical-model approach with a 1000-fold acceleration of spiking neural network dynamics relative to biological real time. Using the embedded plasticity processor, we both simulate the Pong arcade video game and implement a local plasticity rule that enables reinforcement learning, allowing the on-chip neural network to learn to play the game. The experiment demonstrates key aspects of the employed approach, such as accelerated and flexible learning, high energy efficiency and resilience to noise.