AutoML for neuromorphic computing and application-driven co-design: asynchronous, massively parallel optimization of spiking architectures
This work addresses the challenge of application-driven co-design for neuromorphic computing, offering a method to optimize complex spiking architectures, though it appears incremental as it builds on existing AutoML techniques.
The authors tackled the problem of optimizing neuromorphic architectures by extending AutoML approaches with parallel asynchronous model-based search and simulation, achieving efficient exploration of configuration spaces and identification of high-performance conditions for real-time, on-chip learning applications.
In this work we have extended AutoML inspired approaches to the exploration and optimization of neuromorphic architectures. Through the integration of a parallel asynchronous model-based search approach with a simulation framework to simulate spiking architectures, we are able to efficiently explore the configuration space of neuromorphic architectures and identify the subset of conditions leading to the highest performance in a targeted application. We have demonstrated this approach on an exemplar case of real time, on-chip learning application. Our results indicate that we can effectively use optimization approaches to optimize complex architectures, therefore providing a viable pathway towards application-driven codesign.