Negative Feedback Training: A Novel Concept to Improve Robustness of NVCIM DNN Accelerators

arXiv:2305.14561v4 Announce Type: replace
Abstract: Compute-in-memory (CIM) accelerators built upon non-volatile memory (NVM) devices excel in energy efficiency and latency when performing Deep Neural Network (DNN) inference, thanks to their in-situ data processing capability. However, the stochastic nature and intrinsic variations of NVM devices often result in performance degradation in DNN inference. Introducing these non-ideal device behaviors during DNN training enhances robustness, but drawbacks include limited accuracy improvement, reduced prediction confidence, and convergence issues. This arises from a mismatch between …

abstract accelerators arxiv capability compute concept cs.ai cs.ar cs.lg data data processing deep neural network devices dnn dnn accelerators efficiency energy energy efficiency excel feedback however inference in-memory intrinsic latency memory nature negative network neural network novel processing robustness stochastic training type