Jakob Foerster (@j

Jakob Foerster

2,539 posts

Jakob Foerster

@j_foerst

Associate Prof in ML @UniofOxford. Something Something Research Scientist @MetaAI. Something @FLAIR_Ox. Always #teamhuman. Opinions belong to the world.

Joined November 2012

Pinned
Jakob Foerster
@j_foerst
Apr 21, 2025
Making offline RL more honest, reproducible, and robust.
Matthew Jackson
@JacksonMattT
Apr 18, 2025
🌹 Today we're releasing Unifloral, our new library for Offline Reinforcement Learning! We make research easy: ⚛️ Single-file 🤏 Minimal ⚡️ End-to-end Jax Best of all, we unify prior methods into one algorithm - a single hyperparameter space for research! ⤵️
00:00
60K
Jakob Foerster
@j_foerst
Oct 9, 2024
When I discussed quitting Google to do a Phd, my manager, Steve Cheng, gave me the advice of "6 shots": Doing something meaningful usually takes about 5 years and we are productive for roughly 30 years. That gives you 6 attempts. So pick each one carefully and give it your best.
1.4M
Jakob Foerster
@j_foerst
Oct 9, 2024
Replying to @j_foerst
At the meta level, looking back I think it's mindboggling how much positive impact a few minutes of good advice can have. Giving (and listening) to life advice is one of the highest ROI activities ever.
87K
Jakob Foerster
@j_foerst
May 17, 2022
I drafted a quick "How to" guide for writing ML papers. I hope this will be useful (if a little late!) for #NeurIPS2022. Happy paper writing and best of luck!!
docs.google.com
How to ML Paper - A brief Guide
How to ML Paper - A brief Guide Feel free to comment / share and happy paper writing! Also, please see caveats* below. If you like this, why not follow How to ML on Twitter and share the advice/love?...
Jakob Foerster
@j_foerst
Nov 4, 2024
Currently Deep RL is going through an imagenet moment and very few people are aware. This has major implications for RL applications and anyone interested in modeling behaviour (e.g. Econ and neuroscience). To find out more watch my recent talk @ICML2024:
slideslive.com
Jakob Foerster · Reinforcement Learning at the Hyperscale
Deep reinforcement learning is currently undergoing a revolution of scale, fuelled by jointly running the environment, data collection, and training loop on the GPU, which has resulted in orders of...
69K
Jakob Foerster
@j_foerst
Oct 23, 2024
Cold emails are hard and good ones can change a life. Here is my email to @NandoDF that started my career in ML (at the time I was a PM at Google) docs.google.com/document/d/1_u… Real effort (incl feedback) went into drafting it. Thanks to @EugeneVinitsky for nudging me to put it online
docs.google.com
Jakob Foerster's 2015 Cold Email to Nando de Freitas
Subject Follow-up: PhD positions this fall Body Hi Nando, I would like to follow up on the first contact my friend Christoph made today in your Machine Learning lecture at Oxford. I am very interes...
326K
Jakob Foerster
@j_foerst
Sep 21, 2022
My "How to ML Paper - A brief Guide" (docs.google.com/document/d/16R…) is getting visitors again! Good luck with your #ICLR2023 submissions :)
Jakob Foerster
@j_foerst
Jun 29, 2025
I was working at Google before my PhD. But quitting tech to do a PhD allowed me to retool/retrain and to rebrand. Both our skills and how others see them can be limiting factors.
Noam Brown
@polynoamial
Jun 28, 2025
You don’t need a PhD to be a great AI researcher. Even @OpenAI’s Chief Research Officer doesn’t have a PhD.
54K
Jakob Foerster
@j_foerst
Nov 26, 2024
My group at Oxford (@FLAIR_Ox) is talent rich but GPU poor (both compared to industry), so adding more GPUs would be a win for open science, but is difficult to finance from grants. Does anyone have leads for possible donors? Christmas is coming up so I guess I am allow to dream
76K
Jakob Foerster
@j_foerst
Dec 5, 2020
The gradient is a locally greedy direction. Where do you get if you follow the eigenvectors of the Hessian instead? Our new paper, “Ridge Rider” (papers.nips.cc/paper/2020/fil…), explores how to do this and what happens in a variety of (toy) problems (if you dare to do so),.. Thread 1/N
Jakob Foerster
@j_foerst
Dec 14, 2023
LLMs are finally catching up to deep RL - we have been training on test from long before it was cool.
71K
Jakob Foerster
@j_foerst
Sep 11, 2019
Excited to be starting as an Assistant Prof (👨‍🎓!!) at the @UofT (Scarborough Campus) w/ appointment at the @VectorInst in September of 2020. I am looking for exceptional Master/PhD students and Postdocs to be starting with me next fall. Till then, ..
Jakob Foerster
@j_foerst
Oct 1, 2022
"this amounts to solving the multi-agent planning problem" Tesla has now realised that self-driving is a multi-agent problem.. youtu.be/ODSJsviD_SU?t=… 4 years ago I tried to explain to @elonmusk that once CV etc was working, this was the next frontier. He said SL is all you need.
Jakob Foerster
@j_foerst
Nov 12, 2025
FAIRwell, @ylecun. you will be missed.
83K