Matthew T. Jackson

DPhil student at the University of Oxford

matthew_jackson.jpeg

I’m a fourth-year DPhil student at the University of Oxford, where I am co-supervised by Jakob Foerster and Shimon Whiteson. I’m also a member of the AIMS CDT.

I recently completed a research internship at Wayve, working in the World Models Team on video generation for autonomous vehicles. Previously, I completed SWE internships at Amazon, Arm, and Cubica.

Research

Currently, I’m exploring how VLMs can improve RL (synthetic data, representation learning), and how RL can improve VLMs (reasoning, tool-use).

I have published work on offline, meta, and open-ended RL, as well as diffusion models and plasticity.

Policy-Guided Diffusion

Matthew T. Jackson, Michael T. Matthews, Cong Lu, Benjamin Ellis, Jakob Foerster, Shimon Whiteson

RLC 2024 - NeurIPS 2023 Workshop on Robot Learning   paper

Adam on Local Time: Addressing Nonstationarity in RL with Relative Adam Timesteps

Benjamin Ellis*, Matthew T. Jackson*, Andrei Lupu, Alexander D. Goldie, Mattie Fellows, Shimon Whiteson, Jakob N. Foerster

NeurIPS 2024   paper

Discovering Temporally-Aware Reinforcement Learning Algorithms

Matthew T. Jackson*, Chris Lu*, Louis Kirsch, Robert T. Lange, Shimon Whiteson, Jakob N. Foerster

ICLR 2024   paper   podcast

Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design

Matthew T. Jackson, Minqi Jiang, Jack Parker-Holder, Risto Vuorio, Chris Lu, Gregory Farquhar, Shimon Whiteson, Jakob N. Foerster

NeurIPS 2023   paper

Further Publications

Can Learned Optimization Make Reinforcement Learning Less Difficult?

Alexander D. Goldie, Matthew T. Jackson, Chris Lu, Jakob N. Foerster, Shimon Whiteson

NeurIPS 2024 (Spotlight)   paper

Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement Learning

Michael Matthews, Michael Beukman, Benjamin Ellis, Mikayel Samvelyan, Matthew T. Jackson, Samuel Coward, Jakob Foerster

ICML 2024 (Spotlight)   paper

Risks and Opportunities of Open Source Generative AI

Francisco Eiras, Aleksandar Petrov, Bertie Vidgen, Christian Schroeder de Witt, Fabio Pizzati, Katherine Elkins, Supratik Mukhopadhyay, Adel Bibi, Botos Csaba, Fabro Steibel, Fazl Barez, Genevieve Smith, Gianluca Guadagni, Jon Chun, Jordi Cabot, Joseph Marvin Imperial, Juan A. Nolazco-Flores, Lori Landay, Matthew T. Jackson, Paul Röttger, Philip H.S. Torr, Trevor Darrell, Yong Suk Lee, Jakob Foerster

ICML 2024 (Oral)   paper

SplAgger: Split Aggregation for Meta-Reinforcement Learning

Jacob Beck, Matthew T. Jackson, Risto Vuorio, Zheng Xiong, Shimon Whiteson

RLC 2024   paper

Retrieve What You Need: A Mutual Learning Framework for Open-domain Question Answering

Dingmin Wang, Qiuyuan Huang, Matthew T. Jackson, Jianfeng Gao

TACL 2024   paper

Online Reinforcement Learning Controllers for Soft Robots using Learned Environments

Uljad Berdica, Matthew T. Jackson, Jakob Foerster, Perla Maiolino

RoboSoft 2024

Hypernetworks in Meta-Reinforcement Learning

Jake Beck, Matthew T. Jackson, Risto Vuorio, Shimon Whiteson

CoRL 2022   paper

Multi-Modal Fusion by Meta-Initialization

Matthew T. Jackson*, Shreshth A. Malik*, Michael T. Matthews, Yousuf Mohamed-Ahmed

FARSCOPE Robotics Workshop 2022 (Best Poster Award)   paper