hancai [at] mit (dot) edu
I am a second-year Ph.D. student at MIT, advised by Prof. Song Han. Before coming to MIT, I received my Master's and Bachelor's degree at Shanghai Jiao Tong University (SJTU), advised by Prof. Yong Yu. At SJTU, I also worked closely with Prof. Weinan Zhang and Prof. Jun Wang.
My research interests lie in machine learning, particularly efficient deep learning and AutoML, as well as their applications in real-world scenarios such as computer vision, natural language processing, data mining, etc.
- First Place in the CVPR 2020 Workshop of Low-Power Computer Vision Challenge, CPU detection and FPGA track
- First Place in the 2019 IEEE Low-Power Image Recognition Challenge, classification and detection track
- First Place in the Low-Power Computer Vision Workshop at ICCV 2019, DSP track
- Third Place in the CVPR 2019 Workshop Low-Power Image Recognition Challenge, classification track
- Sep 2020: Tiny Transfer Learning: Towards Memory-Efficient On-Device Learning is accepted by NeurIPS'2020.
- Aug 2020: OnceForAll won the first place in the CVPR 2020 Low-Power Computer Vision Challenge, CPU detection and FPGA track.
- May 2020: HAT: Hardware-Aware Transformer for Efficient Natural Language Processing to appear at ACL’2020
- Feb 2020: APQ: Joint Search for Network Architecture, Pruning and Quantization Policy is accepted by CVPR’2020
- Dec 2019: Once-For-All Network (OFA) is accepted by ICLR’2020.
- Nov 2019: AutoML for Architecting Efficient and Specialized Neural Networks to appear at IEEE Micro
- Dec 2018: ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware is accepted by ICLR’2019.
Once for All: Train One Network and Specialize it for Efficient Deployment
OFA is an efficient AutoML technique that decouples model training from architecture search. Train only once, specialize for many hardware platforms, from CPU/GPU to hardware accelerators. OFA consistently outperforms SOTA NAS methods (up to 4.0% ImageNet top1 accuracy improvement over MobileNet-V3) while reducing orders of magnitude GPU hours and CO2 emission. In particular, OFA achieves a new SOTA 80.0% ImageNet top1 accuracy under the mobile setting (<600M FLOPs). OFA is the winning solution for CVPR 2020 Workshop of Low-Power Computer Vision Challenge (FPGA track), 2019 IEEE Low-Power Image Recognition Challenge (classification and detection track), Low-Power Computer Vision Workshop at ICCV 2019 (DSP track). [Media: MIT News, Qualcomm News, VentureBeat][Code: GitHub (800+ stars), Colab]
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
ProxylessNAS is an efficient hardware-aware neural architecture search method, which can directly search on large-scale datasets (e.g., ImageNet). ProxylessNAS is hardware-aware. It can design specialized neural network architecture for different hardware platforms. With >74.5% top-1 accuracy, the latency of ProxylessNAS is 1.8x faster than MobileNetV2, the current industry standard for mobile vision. [Media: MIT News, IEEE Spectrum][Industry Integration: PytorchHub@Facebook, AutoGluon@AWS][Code: GitHub (1.1K stars)]
- Serve as a reviewer for TPAMI, ICML 2020, NeurIPS 2020/2019, ICLR 2021