Greetings! I am a graduate student at Department of Computer and Information Science, University of Pennsylvania. Currently, I am working on event representation (a sub-area in NLP) in Cognitive Computation Group, following Prof. Dan Roth.
I was an undergrad at Shanghai Jiao Tong University (上海交通大学), majoring in Information Engineering from 2015 to 2019. My undergraduate thesis was on traffic optimization in datacenter networks with deep reinforcement learning under the supervision of Prof. Lingge Jiang. I was a visiting student of the Natural Language Processing group at UCSB in the summer of 2018, advised by Prof. William Wang and mentored by Dr. Vivek Kulkarni. I did most of my undergrad research at the Machine Vision and Intelligence Group at SJTU, advised by Prof. Cewu Lu. I was also a research assistant at the National Engineering Laboratory for Information Content Analysis Technology at SJTU, advised by Prof. Shenghong Li.
I am interested in applying machine learning algorithms to problems in Natural Language Processing, Computer Vision, and Computer Networks. My ultimate academic goal is to design algorithms that are as cognitively capable as human beings in every meaningful dimensions.
Human Pose Tracking: A new method for multi-person pose tracking with spatio-temporal information
6D Object Pose Estimation: A novel architecture of detecting 3D model instances and estimating 6D pose under occlusion
Knowledge Graph Embedding: A knowledge graph embedding algorithm that captures contextual cues and dependencies among entities and relations
Traffic Optimization with Deep Reinforcement Learning: A hierarchical architecture for load balancing and flow scheduling in datacenter networks.
1. Pose Flow: Efficient Online Pose Tracking, BMVC 2018
Yuliang Xiu *, Jiefeng Li *, Haoyu Wang, Yinghong Fang, and Cewu Lu. [code]
2. Estimating 6D Pose From Localizing Designated Surface Keypoints, CVPR 2019 submission
Zelin Zhao, Haoyu Wang *, Gao Peng *, Haoshu Fang, Chengkun Li, and Cewu Lu. [code]
3. DOLORES: Deep Contextualized Knowledge Graph Embeddings, EMNLP 2019 submission
Haoyu Wang, Vivek Kulkarni, and William Yang Wang. [code]
4. A Study on Load Balancing Algorithm with Lightweight AI, Undergraduate Thesis
Haoyu Wang and Lingge Jiang.
(* indicates equal contribution)
EE Summer Intensive Program, University of Washington, Seattle, USA, 2016.6
Winter International Student Experience, Sungkyunkwan University (성균관대학교), Seoul, South Korea, 2019.1
Piano: started to play at the age of 4
Tennis: served as President of Tennis Association at SJTU from 2016.6 to 2017.6