Bespoke Machine Learning Processor Development Framework on Flexible Substrates

Abstract

This paper proposes a framework for the design of bespoke machine learning (ML) processors on flexible substrates (e.g. plastic) to address an important need in flexible and wearable applications – a processing engine of the flexible electronics applications. The proposed framework automates the design of bespoke ML processors on flexible substrates to reduce development time, and therefore the time-to-market.