Jiayuan Mao (茅佳源)

I am an undergraduate student at IIIS (YaoClass), Tsinghua University, studying structured knowledge representation and reasoning. By idenfying and trying to tackle the problems lying between the gap of human and machine intelligence, I explore and exploit the structures of knowledge/memory and the inductive biases for reasoning.

I am a visiting student of COCOSCI group at MIT, adviced by Prof. Josh Tenenbaum. I was a visiting student of CLVR group at USC during the spring and summer of 2017, adviced by Prof. Joseph Lim. I spent most of my undergraduate years being an intern researcher at Megvii, Inc., where my research career began. I also had a great time at Google AI China Center and Bytedance Inc.

Research topics
  • Neural abstraction machines [?]: neural inductive logic [NIPS18sub], structured memory augmented neural networks [ICLR18].
  • Transfer learning and life-long learning: task transfer and generalization in DRL [ICLR18], relationship between tasks in vision systems [CVPR17].
  • Concept Learning: visual-semantic representation [COLING18].
  • Sentential semantic representation.
  • Object detection: [CVPR17] [ECCV18].

Publications (by date)

(* indicates equal contribution)

Acquisition of Localization Confidence for Accurate Object Detection
Borui Jiang*, Ruixuan Luo*, Jiayuan Mao*, Tete Xiao, Yuning Jiang

ECCV 2018 (Oral) [Arxiv] [Code] [Webpage]

Learning Visually-Grounded Semantics from Contrastive Adversarial Samples
Haoyue Shi*, Jiayuan Mao*, Tete Xiao*, Yuning Jiang, Jian Sun

COLING 2018 [Arxiv] [Code] [Webpage]

Universal Agent for Disentangling Environments and Tasks
Jiayuan Mao, Honghua Dong, Joseph J. Lim

ICLR 2018 [OpenReview] [Webpage]

What Can Help Pedestrian Detection?
Jiayuan Mao*, Tete Xiao*, Yuning Jiang, Zhimin Cao

CVPR 2017 [ArXiv] [Webpage]