Biography
Yangcheng Gao is currently a MS student of Hefei University of Technology (HFUT), majoring in Computer Science, under the supervision of Zhao Zhang. He has worked in Shanghai AI Laboratory as a research intern, from April to July 2022. Previously he recieved a BE degree in Automotive Engineering from HFUT.
He focuses on the theories and algorithms for DNN model accelaration, model deployment and model efficiency. His research interests include low-bit quantization and tensor decomposition. Besides, he also focuses on the applications of DNN on robotics.
News
- [ 26 September 2022] Congratulations! Yangcheng Gao received National Scholarship.
- [ 15 September 2022] ECCV MIPI 2022 Challenge on Under-Display Camera Image Restoration results are released now. Our team got Fifth Place!
- [ 1 September 2022 ] The paper “Towards Feature Distribution Alignment and Diversity Enhancement for Data-Free Quantization” has been accepted by IEEE ICDM 2022.
- [ 31 August 2021 ] The paper “Dictionary Pair-based Data-Free Fast Deep Neural Network Compression” has been accepted by IEEE ICDM 2021, and invited for KAIS journal publication as “Best-ranked” paper.
Publications
Yan Luo#, Yangcheng Gao#, et al. “Long-Range Zero-Shot Generative Deep Network Quantization,” submitted to CVPR, 2023.
Zhao Zhang, Yangcheng Gao, et al. “SelectQ: Calibration Data Selection for Post-Training Quantization,” submitted to CVPR, 2023.
…, Huan Zheng, Yangcheng Gao, et al. “MIPI 2022 Challenge on Under-Display Camera Image Restoration: Methods and Results,” Mobile Intelligent Photography and Imaging Workshop 2022 (ECCV MIPI), Tel-Aviv, Israel, Oct 2022.
Yangcheng Gao, Zhao Zhang, et al. “ClusterQ: Semantic Feature Distribution Alignment for Data-Free Quantization,” submitted to IEEE TNNLS, 2022.
Yangcheng Gao, Zhao Zhang, et al. “Towards Feature Distribution Alignment and Diversity Enhancement for Data-Free Quantization,” IEEE ICDM 2022.
Yangcheng Gao, Zhao Zhang, et al. “Fast and Effective Data-Free Deep Neural Network Compression by Dictionary Pair-Driven Reconstruction,” submitted to KAIS, 2022.
Yangcheng Gao, Zhao Zhang, et al. “Dictionary Pair-based Data-Free Fast Deep Neural Network Compression,” IEEE ICDM 2021. (Invited for KAIS journal publication as Best-ranked paper)