C-BRIC Arizona State Researchers Win TVLSI Best Paper Award

Congratulations to C-BRIC researchers Shihui Yin, Minkyu Kim, Tushar Gupta (all C-BRIC graduates), and Jae-sun Seo of Arizona State University.

 

Yin, Kim, Gupta, Seo, and collaborators won the Best Paper Award for the IEEE Transactions on Very Large Scale Integration Systems (TVLSI). TVLSI "covers system specification, design and partitioning, system-level test, and reliable VLSI/ULSI systems."

 

C-BRIC researchers, along with Zhewei Jiang and Mingoo Seok of Columbia University, earned the award for their work on “Vesti: Energy-Efficient In-Memory Computing Accelerator for Deep Neural Networks.”  In this work, the group developed a new DNN accelerator: Vesti. Vesti is "designed to support configurable multi-bit activations and large-scale DNNs seamlessly while substantially improving the chip-level energy-efficiency with favorable accuracy trade-off, compared to conventional digital ASIC. Vesti also employs double-buffering with two groups of in-memory computing SRAMs, effectively hiding the row-by-row write latencies of in-memory computing SRAMs." Vesti demonstrates ultra-low energy consumption.