Papers in the reversed chronological order. Generated by jekyll-scholar. For the latest publication, please see the Google Scholar.

Notation: * indicates equal contribution, ° indicates equal advising.
Abbreviations for Field: CV for Computer Vision, HRI for Human-Robot Interaction, RL for Reinforcement Learning.
Each paper is captioned by a TLDR summary.


  1. AAAI-W
    Adaptive Agent Architecture for Real-Time Human-Agent Teaming
    Tianwei Ni*, Huao Li*, Siddharth Agrawal*, Suhas Raja, Fan Jia, Yikang Gui, Dana Hughes, Michael Lewis, and Katia Sycara
    AAAI workshop on Plan, Activity, and Intent Recognition (PAIR), 2021

    Field: HRI. TLDR: adaptive agent to human teammates by self-play team performance projection.


  1. CoRL
    f-IRL: Inverse Reinforcement Learning via State Marginal Matching
    In Conference on Robot Learning (CoRL), 2020

    Field: RL. TLDR: learn state-only reward functions for downstream tasks. Abridged version in RSS workshop.

  2. ICML-W
    Meta-SAC: Auto-tune the Entropy Temperature of Soft Actor-Critic via Metagradient
    Yufei Wang*, and Tianwei Ni*
    7th ICML Workshop on Automated Machine Learning (AutoML), 2020

    Field: RL. TLDR: meta-learn the entropy temperature for exploration-exploitation trade-off.

  3. HFES
    Team Synchronization and Individual Contributions in Coop-Space Fortress
    64th Annual Meeting of Human Factors and Ergonomics Society (HFES), 2020

    Field: HRI. TLDR: the connection betweeen team performance and sychronization in human teams.


  1. CVPR
    Elastic Boundary Projection for 3D Medical Image Segmentation
    Tianwei Ni, Lingxi Xie, Huangjie Zheng, Elliot K Fishman, and Alan Yuille
    In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019

    Field: CV. TLDR: reformulate object segmentation as behavior cloning problem.


  1. arXiv
    Phase Collaborative Network for Two-Phase Medical Image Segmentation
    Huangjie Zheng, Lingxi Xie, Tianwei Ni, Ya Zhang, Yan-Feng Wang, Qi Tian, Elliot K Fishman, and Alan Yuille
    Preprint, 2018

    Field: CV. TLDR: CycleGAN for unpaired two-phase medical image segmentation.