A 28-nm 8-bit Floating-Point Tensor Core-Based Programmable CNN Training Processor With Dynamic Structured Sparsity.
Shreyas Kolala VenkataramanaiahJian MengHan-Sok SuhInjune YeoJyotishman SaikiaSai Kiran CherupallyYichi ZhangZhiru ZhangJae-Sun SeoPublished in: IEEE J. Solid State Circuits (2023)