Before that, I conducted my undegraduate thesis in VIPL Group at Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS), under the supervision of Prof. Xilin Chen and Prof. Xiujuan Chai. Before that, I interned in Human Motion Research Group at ICT, CAS, supervised by Prof. Lin Gao, Prof. Yu-Kun Lai and Prof. Shihong Xia. I also have serveral research experiences in interdisciplinary science, including UROP projects in Institute of Medical Engineering & Science and MIT Sloan School of Management of MIT, and synthetic biology project for the International Genetically Engineered Machine (iGEM) competition.
Ph.D. Student in Computer Science, 2018-present
University of Maryland, College Park
B.Eng. in Computer Science and Technology, 2014-2018
University of Chinese Academy of Sciences
Special Student in EECS, 2017
Massachusetts Institute of Technology
We present a novel algorithm (DeepMNavigate) for global multi-agent navigation in dense scenarios using deep reinforcement learning. Our approach uses local and global information for each robot based on motion information maps. We demonstrate the performance on complex, dense benchmarks with narrow passages on environments with tens of agents. We highlight the algorithm’s beneﬁts over prior learning methods and geometric decentralized algorithms in complex scenarios.
We address the problem of accelerating thin-shell deformable object simulations by dimension reduction. We present a new algorithm to embed a high-dimensional conﬁguration space of deformable objects in a low-dimensional feature space, where the conﬁgurations of objects and feature points have approximate one-to-one mapping.
We propose a novel framework which we call mesh variational autoencoders, to explore the probabilistic latent space of 3D surfaces. The framework is easy to train, and requires very few training examples. We also propose an extended model which allows flexibly adjusting the significance of different latent variables by altering the prior distribution.
We propose to combine the identification of key temporal and spatial regions and sign language classification into one deep learning network architecture, using attention mechanisms to focus on effective features, and realizing end-to-end automatic sign language recognition. Our proposed architecture can enhance the efficiency of sign language recognition and maintain a comparable recognition accuracy with state-of-art work.
We propose a novel mesh-based autoencoder architecture that is able to cope with meshes with irregular topology. We introduce sparse regularization in this framework, which along with convolutional operations, helps localize mesh deformations. Our framework is capable of extracting localized deformation components from mesh data sets with large-scale deformations and is robust to noise.