Evaluating complexity and resilience trade-offs in emerging memory inference machines

2020/02/25 by Christopher R. Bennett, Bennett, Christopher H., Ryan Dellana +9
Engineering · Computer Science · #Advanced Memory and Neural Computing #Ferroelectric and Negative Capacitance Devices #Neural Networks and Applications

paper · pdf · doi:10.48550/arxiv.2003.10396

Abstract

Neuromorphic-style inference only works well if limited hardware resources are maximized properly, e.g. accuracy continues to scale with parameters and complexity in the face of potential disturbance. In this work, we use realistic crossbar simulations to highlight that compact implementations of deep neural networks are unexpectedly susceptible to collapse from multiple system disturbances. Our work proposes a middle path towards high performance and strong resilience utilizing the Mosaics framework, and specifically by re-using synaptic connections in a recurrent neural network implementation that possesses a natural form of noise-immunity.

Related

• Training Recurrent Neural Networks against Noisy Computations during Inference
• On the Mitigation of Read Disturbances in Neuromorphic Inference Hardware
• On The Robustness of a Neural Network
• Noisy Computations during Inference: Harmful or Helpful?
• Improving Inference Lifetime of Neuromorphic Systems via Intelligent Synapse Mapping