Jake Grigsby

profile_pic.jpeg

I am a second year CS PhD student at UT Austin, working with Prof. Yuke Zhu and the Robot Perception and Learning Lab. My research focuses on generalization and long-term memory in deep reinforcement learning. Before coming to Austin, I studied Math and CS at the University of Virginia, where my research was advised by Prof. Yanjun Qi.

Research

  1. AMAGO: Scalable In-Context Reinforcement Learning for Adaptive Agents
    Grigsby, Jake, Fan, Jim,  and Zhu, Yuke
    ICLR (Spotlight) 2024
    PDF Code Website
  2. Long-Range Transformers for Dynamic Spatiotemporal Forecasting
    Grigsby, Jake, Wang, Zhe, Nguyen, Nam,  and Qi, Yanjun
    KDD Workshop on Mining and Learning from Time Series 2023 (Released 2021)
    PDF Code
  3. Cross-Episodic Curriculum for Transformer Agents
    Shi, Lucy Xiaoyang, Jiang, Yunfan,  Grigsby, Jake, Fan, Jim,  and Zhu, Yuke
    NeurIPS 2023
    PDF Website
  4. PGrad: Learning Principal Gradients For Domain Generalization
    Wang, Zhe,  Grigsby, Jake,  and Qi, Yanjun
    ICLR 2022
    PDF
  5. ST-MAML: A Stochastic-Task based Method for Task-Heterogeneous Meta-Learning
    Wang, Zhe,  Grigsby, Jake, Sekhon, Arshdeep,  and Qi, Yanjun
    Conference on Uncertainty in Artificial Intelligence 2022
    PDF
  6. RARE: Renewable Energy Aware Resource Management in Datacenters
    Venkataswamy, Vanamala,  Grigsby, Jake, Grimshaw, Andrew,  and Qi, Yanjun
    Workshop on Job Scheduling for Parallel Processing 2022
    PDF
  7. Towards Automatic Actor-Critic Solutions to Continuous Control
    Grigsby, Jake, Yoo, Jin Yong,  and Qi, Yanjun
    NeurIPS Workshop on Deep Reinforcement Learning 2021
    PDF Code
  8. A Closer Look at Advantage-Filtered Behavioral Cloning in High-Noise Datasets
    Grigsby, Jake,  and Qi, Yanjun
    UVA Distinguished Major Thesis 2021
    PDF
  9. Deep learning analysis of deeply virtual exclusive photoproduction
    Grigsby, Jake, Kriesten, Brandon, Hoskins, Joshua, Liuti, Simonetta,  Alonzi, Peter and 1 more author
    Phys. Rev. D 2021
    PDF
  10. Measuring Visual Generalization in Continuous Control From Pixels
    Grigsby, Jake,  and Qi, Yanjun
    NeurIPS Workshop on Deep Reinforcement Learning 2020
    PDF Code