SHF: Small: Fast Sign-Off of Machine Learning Systems: From Circuit-Level Modeling to Statistical System Validation

Duke University, Durham NC

Investigators

Abstract

Machine learning has been adopted by a broad range of emerging applications, including health monitoring, autonomous driving, advanced manufacturing, etc. However, any machine learning system cannot be 100% accurate due to the accuracy limitation posed by machine learning algorithms and the circuit-level non-ideal features associated with its hardware implementation. This project investigates a radically new framework for efficient validation of machine learning systems implemented with nano-scale integrated circuits. It aims to identify and synthesize the critical corner cases for which a machine learning system is likely to fail. The project is expected to initialize a paradigm shift in today's design methodology for complex machine learning systems, thereby leading to an immediate impact on a broad range of industrial sectors relying on machine intelligence. In addition, the proposed education activities create a large number of unique training opportunities for both academic and industrial participants, substantially improving the education infrastructure and generate high-quality researchers and practitioners for the society. Today, validating a machine learning system with high throughout, low power and complex functionality is an extremely challenging task. This project attacks the grand challenge by developing a novel validation framework composed of two major components: (1) corner-case generation and (2) rare-failure rate estimation. Both physical circuit models and statistical generative models are proposed to synthesize a large amount of test cases, reducing the experimental cost to physically record the corner-cases that are difficult to observe. Furthermore, a novel formulation, based on subset partition and graph embedding, is developed to efficiently inspect the likely-failed test cases and consequently estimate the rare- failure rate that is expensive to capture by random sampling. Built upon these mathematical tools, the project's framework offers a fundamental infrastructure that could facilitate radical breakthroughs over numerous machine learning applications. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.