Intro

Currently I am a PhD student with Prof. Foerster J. at FLAIR at the University of Oxford and previously Vector Institute, University of Toronto. During my PhD, I was a Research Intern at Google DeepMind in the Montreal team. Before my PhD, I was a Research Scientist at Nnaisense, where we developed Deep Learning based technologies to solve problems in industry. Before joining Nnaisense, I completed my MSc in Artificial Intelligence at the Università della Svizzera italiana.

My research interests include Multi-Agent Reinforcement Learning, Generative Models and Meta-Learning.

Highlighted Work

Mixture of Experts in a Mixture of RL settings, RLC 2024
Willi* T., Obando-Ceron* J., Foerster J.,Dziugaite K., Castro P.

Mixtures of Experts Unlock Parameter Scaling for Deep RL, ICML 2024 (spotlight, 3.5%)
Obando-Ceron* J., Sokar* G., Willi* T., Lyle C., Farebrother J., Foerster J., Dziugaite K., Precup D., Castro P.

JaxMARL: Multi-Agent RL Environments and Algorithms in JAX, AAMAS 2024, (ext. abs)
Rutherford* A., Ellis* B., Gallici* M., Cook* J., Lupu* A., Ingvarsson* G., Willi* T., Khan A., Schroeder de Witt C., Souly A., Bandyopadhyay S., Samvelyan M, Jiang M., Lange R., Whiteson S., Lacerda B., Hawes N., Rocktäschel T., Lu* C., Foerster J.

Analysing the Sample Complexity of Opponent Shaping, AAMAS 2024 (oral)
Fung K., Zhang Q., Lu C., Willi T., Foerster J.

Leading the Pack: N-player Opponent Shaping, Workshop @ Neurips 2023 (oral)
Souly A.,Willi T.,Khan A., Kirk R., Lu C., Grefenstette E., Rocktäschel T.

Pax: Scalable Opponent Shaping in Jax, Github
Willi T., Khan* A., Kwan* N., Lu C., Foerster J.

Scaling Opponent Shaping to High Dimensional Games, AAMAS 2024 (oral)
Khan* A., Willi* T., Kwan* N.,Tacchetti A.,Lu C., Grefenstette E., Rocktäschel T., Foerster J.

Adversarial Cheap Talk, ICML 2023 & 5 Workshops @ ICML 2022
Lu C., Willi T., Letcher A., Foerster J.

Model-Free Opponent Shaping, ICML 2022 (short presentation, 22%)
Lu C., Willi T., Schroeder de Witt C., Foerster J.

COLA: Consistent Learning with Opponent-Learning Awareness, ICML 2022 (short presentation, 22%)
Willi T.*, Treutlein J.*, Letcher A.*, Foerster J.

Recurrent Neural Processes, Neurips 2019 BDL Workshop (oral)
Willi T., Masci J., Schmidhuber J., Osendorfer C.

Master's Thesis "Recurrent Neural Processes"
Supervised by Schmidhuber J. and Osendorfer C.

Miscellaneous

Talks

Races

Date Event Time Distance Ascent
22.10.2023 Transviamala 02:12:08 19.9km +950m/-620m
21.10.2023 Transruinaulta Trailmarathon 06:27:50 42.2km 1800m
10.6.2023 Exmoor Trailultra 07:42:56 52.4km 2068m
14.5.2023 Oxford Town and Gown 47:29 10km 0m
18.6.2022 Oxford to Cambridge (DNF) 30h 105km ??
8.5.2022 Oxford Town and Gown 47:17 10km 0m
6.3.2022 Cambridge Halfmarathon 1:55:38 21.1km 0m
9.12.2021 Inofficial Timon's Oxford Marathon 5:02:18 42.38km 250m
21.8.2021 Greinatrail 4:38:07 27.38km 1411m
20.10.2019 Transviamala 2:33:20.0 19km +950m/-620m
19.10.2019 Transruinaulta Trailmarathon 5:46:05.8 42.2km 1800m
15.6.2019 Scenic Trail K54 (DNF) 9:51:02 50km 3566m
21.10.2018 Transviamala 2:12:12.9 19km +950m/-620m
20.10.2018 Transruinaulta Trailmarathon 6:30:00.7 42.2km 1800m
9.6.2018 Scenic Trail K27 4:22:17 27km 2100m
22.10.2017 Transviamala 2:08:33.7 19km +950m/-620m
21.10.2017 Transruinaulta Trailmarathon 5:28:15.9 42.2km 1800m
1.1.2017 Neujahrshalbmarathon Schlieren 2:02:42.4 21.1km 0m
23.10.2016 Transviamala 2:11:17.3 19km +950m/-620m
19.10.2014 Transviamala 2:23:44.2 19km +950m/-620m

Elements

Text

This is bold and this is strong. This is italic and this is emphasized. This is superscript text and this is subscript text. This is underlined and this is code: for (;;) { ... }. Finally, this is a link.


Heading Level 2

Heading Level 3

Heading Level 4

Heading Level 5
Heading Level 6

Blockquote

Fringilla nisl. Donec accumsan interdum nisi, quis tincidunt felis sagittis eget tempus euismod. Vestibulum ante ipsum primis in faucibus vestibulum. Blandit adipiscing eu felis iaculis volutpat ac adipiscing accumsan faucibus. Vestibulum ante ipsum primis in faucibus lorem ipsum dolor sit amet nullam adipiscing eu felis.

Preformatted

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';

Lists

Unordered

  • Dolor pulvinar etiam.
  • Sagittis adipiscing.
  • Felis enim feugiat.

Alternate

  • Dolor pulvinar etiam.
  • Sagittis adipiscing.
  • Felis enim feugiat.

Ordered

  1. Dolor pulvinar etiam.
  2. Etiam vel felis viverra.
  3. Felis enim feugiat.
  4. Dolor pulvinar etiam.
  5. Etiam vel felis lorem.
  6. Felis enim et feugiat.

Icons

Actions

Table

Default

Name Description Price
Item One Ante turpis integer aliquet porttitor. 29.99
Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
Item Three Morbi faucibus arcu accumsan lorem. 29.99
Item Four Vitae integer tempus condimentum. 19.99
Item Five Ante turpis integer aliquet porttitor. 29.99
100.00

Alternate

Name Description Price
Item One Ante turpis integer aliquet porttitor. 29.99
Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
Item Three Morbi faucibus arcu accumsan lorem. 29.99
Item Four Vitae integer tempus condimentum. 19.99
Item Five Ante turpis integer aliquet porttitor. 29.99
100.00

Buttons

  • Disabled
  • Disabled

Form