Zhi Gao (高志)
I am a postdoctoral research fellow at the School of Intelligence Science and Technology, Peking University, working with Prof. Song-Chun Zhu. (北京大学智能学院)
Now, I am also a visiting researcher at the machine learning group at Beijing Institute for General Artificial Intelligence (北京通用人工智能研究院). We are recruiting interns and full-time staff.
E-mail: zhi.gao@pku.edu.cn, gaozhi_2017@bit.edu.cn, and gaozhi_2017@126.com
Education
I received my B.S. degree in computer science from the Beijing Institute of Technology (BIT), Beijing, China, in 2017.
I received my Ph.D. degree in computer science from the Beijing Institute of Technology (BIT), Beijing, China, in 2023, under the supervision of Prof. Yuwei Wu and Prof. Yunde Jia.
Research Interest
My research interests lie in computer vision and machine learning. Concretely,
(1) I am interested in multimodal agents. Recently, I am working on powerful multimodal models, closed-loop frameworks, and feedback learning, through which agents could learn from human-robot interaction and fast adapt to new environments.
(2) I am interested in structure and geometry of data space. In my graduate studies, I work on learning representations on Riemannian manifolds to match some properties of data.
(3) I am interested in machine learning in the open environment where insufficient annotation is provided and there is distribution shift between data, such as few-shot learning, meta-learning, self-supervised learning, transfer learning, and continual learning.
Publications
2024
高志、武玉伟、贾云得. 混合曲率空间中的几何自适应元学习方法. 计算机学报 2024.
Yue Fan, Xiaojian Ma, Rujie Wu, Yuntao Du, Jiaqi Li, Zhi Gao, Qing Li. VideoAgent: A Memory-augmented Multimodal Agent for Video Understanding. ECCV 2024.
Zhi Gao, Yuntao Du, Xintong Zhang, Xiaojian Ma, Wenjuan Han, Song-Chun Zhu, Qing Li. CLOVA: A Closed-LOop Visual Assistant with Tool Usage and Update. CVPR 2024. [pdf]
2023
My Thesis. Geometry-Adaptive Meta-Learning in Riemannian Manifolds. (in Chinese) [pdf]
Zhi Gao, Chen Xu, Feng Li, Yunde Jia, Mehrtash Harandi, Yuwei Wu. Exploring Data Geometry for Continual Learning. CVPR 2023. [pdf]
Jin Chen+, Zhi Gao+, Xinxiao Wu, Jiebo Luo. Meta-causal Learning for Single Domain Generalization. CVPR 2023. (co-first author) [pdf]
2022
Zhi Gao, Yuwei Wu, Xiaomeng Fan, Mehrtash Harandi, Yunde Jia. Learning to Optimize on Riemannian Manifolds. T-PAMI 2022. [pdf]
Zhi Gao, Yuwei Wu, Mehrtash Harandi, Yunde Jia. Curvature-Adaptive Meta-Learning for Fast Adaptation to Manifold Data. T-PAMI 2022. [pdf]
Zhi Gao, Yuwei Wu, Yunde Jia, Mehrtash Harandi. Hyperbolic Feature Augmentation via Distribution Estimation and Infinite Sampling on Manifolds. NeurIPS 2022. [pdf]
Xiaomeng Fan, Yuwei Wu, Zhi Gao, Yunde Jia, Mehrtash Harandi. Efficient Riemannian Meta-Optimization by Implicit Differentiation. AAAI 2022. [pdf]
2021
Zhi Gao, Yuwei Wu, Yunde Jia, Mehrtash Harandi. Curvature Generation in Hyperbolic Spaces for Few-Shot Learning. ICCV 2021. [pdf]
Xiaomeng Fan+, Zhi Gao+, Yuwei Wu, Yunde Jia, Mehrtash Harandi. Learning a Gradient-free Riemannian Optimizer on Tangent Spaces. AAAI 2021. (co-first author) [pdf]
Jindou Dai, Yuwei Wu, Zhi Gao, Yunde Jia. A Hyperbolic-to-Hyperbolic Graph Convolutional Network. CVPR 2021 (oral). [pdf]
Jindou Dai, Yuwei Wu, Zhi Gao, Yunde Jia. Infinite-dimensional Feature Aggregation via a Factorized Bilinear Model. Pattern Recognition 2021. [pdf]
2020
Zhi Gao, Yuwei Wu, Yunde Jia, Mehrtash Harandi. Learning to Optimize on SPD Manifolds. CVPR 2020. [pdf]
Zhi Gao, Yuwei Wu, Xiaoxun Zhang, Jindou Dai, Yunde Jia, Mehrtash Harandi. Revisiting Bilinear Pooling: A Coding Perspective. AAAI 2020. [pdf]
2019
Zhi Gao, Yuwei Wu, Mehrtash Harandi, Yunde Jia. A Robust Distance Measure for Similarity-Based Classification on the SPD Manifold. TNNLS 2019. [pdf]
Zhi Gao, Yuwei Wu, Xingyuan Bu, Tan Yu, Junsong Yuan, Yunde Jia. Learning a robust representation via a deep network on symmetric positive definite manifolds. Pattern Recognition 2019. [pdf]
Xingyuan Bu, Yuwei Wu, Zhi Gao, Yunde Jia. Deep convolutional network with locality and sparsity constraints for texture classification. Pattern Recognition 2019. [pdf]
Reward
ACM SIGAI China Doctoral Dissertation Award 2023