Kaisheng Ma

TL;DR: A general training framework named self distillation, which notably enhances the performance of convolutional neural networks through shrinking the size of the network rather than aggrandizing it, and can also provide flexibility of depth-wise scalable inference on resource-limited edge devices.

TL;DR: The simulation platform in this paper is calibrated using measured results from a fabricated nonvolatile processor and used to explore the design space for a nonVolatile processor with different architectures, different input power sources, and policies for maximizing forward progress.

TL;DR: New metrics of nonvolatile processors to consider energy harvesting factors for the first time are proposed and the nonvolatility processor design from circuit to system level is explored.

TL;DR: This work proposes a 2-transistor (2T) FEFET-based nonvolatile memory with separate read and write paths that achieves non-destructive read and lower write power at iso-write speed compared to standard FE-RAM.

TL;DR: The authors proposed a framework of concurrent adversarial training and weight pruning that enables model compression while still preserving the adversarial robustness and essentially tackles the dilemma of adversarial learning, which is well known that deep neural networks are vulnerable to adversarial attacks, which are implemented by adding crafted perturbations onto benign examples.