Energy Efficient Computing
- Neural Network Accelerator Comparison: A comprehensive summary chart of neural network accelerators. Contributions and suggestions are welcome!
- Nov., 2019 Prof. Yu Wang gave a keynote talk at ASICON. asicon-talk (24.2 MB)
- Feb., 2016 Dr. Xiaoming Chen (co-advised by Prof. Huazhong Yang and Dr. Yu Wang) received 2015 EDAA Outstanding Dissertation Award, which is the first awardee in mainland China in the history of this award.
- Jan., 2016 Dr. Yu Wang is selected into the ACM Distinguished Speaker Program from 2016 to 2018 (3 years). detail
- Nov., 2015 A-Eye, the smart camera was listed as highlight in innovations of 2015 by Microsoft Research Asia detail
- Jul. 2020, Update on Neural Network Accelerator Comparison (Slide with voice)
Feature size of the CMOS transistor keeps going down to reduce the area and power of integrated circuits. A lot of side effects come along with the scaling down, how to build up a reliable and low power system based on un-reliable nano-scale devices becomes a critical problem. We focus on three main topics:
- Leakage power and reliability (especially aging) aware design methodology
- Fast/Parallel Circuit Simulation
- Heterogenous Integration for 3D IC.
Integration of more processing element and memory is another way to integrate more transistors, so that one single IC can have more functions. However, how to map different applications to multi/many core system or directly to transistors (by FPGA or ASIC), and then make these silicon work in a more efficient way bring us opportunities to research in the application specific hardware computing area. We mainly focus on the basic key operations: matrix operations, graph theoretical algorithms, and etc. We category our research according to different applications or computing frameworks
We look into the neuroscience area (mainly about Brain Imaging), and using our hardware computing techniques to help the neuroscientists and doctors to reveal more interesting insights from the Imaging based techniques (XMRI).
On the other hand, we are paying more attention to new devices based emerging computing system. We are considering to use RRAM/Memristor to build new circuit components, which can reduce orders of computational complexity in a different layer comparing with the algorithm optimization for some specific application domains.