Antonin Raffin
@araffin.bsky.social
Researcher in robotics and machine learning (Reinforcement Learning). Maintainer of Stable-Baselines (SB3).
https://araffin.github.io/
https://araffin.github.io/
Pinned
Antonin Raffin
@araffin.bsky.social
· Oct 24
Post your most popular 🐦 from Twitter
Types of Reinforcement Learning Paper
Original image: @xkcd.com
Types of Reinforcement Learning Paper
Original image: @xkcd.com
Reposted by Antonin Raffin
The talk i gave about "Recent Advances in RL for Continuous Control" at CERN last year is now online =)
www.youtube.com/watch?v=Sb0d...
www.youtube.com/watch?v=Sb0d...
Recent Advances in RL for Continuous Control (SOTA 2025) | CERN ML Workshop
YouTube video by Antonin Raffin
www.youtube.com
February 11, 2026 at 7:05 AM
The talk i gave about "Recent Advances in RL for Continuous Control" at CERN last year is now online =)
www.youtube.com/watch?v=Sb0d...
www.youtube.com/watch?v=Sb0d...
Reposted by Antonin Raffin
nice blog post about a humanoid robotics startup failure: ruixu.us/posts/six-th...
Six Things I Learned Watching a Robotics Startup Die from the Inside | Rui Xu
I spent a year as COO of a YC-backed robotics startup. The company didn't make it. Here's what I actually learned.
ruixu.us
February 11, 2026 at 2:04 PM
nice blog post about a humanoid robotics startup failure: ruixu.us/posts/six-th...
The talk i gave about "Recent Advances in RL for Continuous Control" at CERN last year is now online =)
www.youtube.com/watch?v=Sb0d...
www.youtube.com/watch?v=Sb0d...
Recent Advances in RL for Continuous Control (SOTA 2025) | CERN ML Workshop
YouTube video by Antonin Raffin
www.youtube.com
February 11, 2026 at 7:05 AM
The talk i gave about "Recent Advances in RL for Continuous Control" at CERN last year is now online =)
www.youtube.com/watch?v=Sb0d...
www.youtube.com/watch?v=Sb0d...
Reposted by Antonin Raffin
If you missed this post last week, it explains pretty well how modern frontend works these days. :/
https://paulmakeswebsite...
https://paulmakeswebsite...
February 2, 2026 at 11:45 AM
If you missed this post last week, it explains pretty well how modern frontend works these days. :/
https://paulmakeswebsite...
https://paulmakeswebsite...
Q-value overestimation animation for my upcoming talk about "Recent Advances in RL for Continuous Control" at the Mannheim RL Workshop
January 31, 2026 at 1:41 PM
Q-value overestimation animation for my upcoming talk about "Recent Advances in RL for Continuous Control" at the Mannheim RL Workshop
Reposted by Antonin Raffin
This is something I talk about in my paper, where I suggest being explicit about {\gamma}_train (some methods use multiple gammas during training) and \gamma_eval.
One of my students is empirically investigating this and, as one would expect, it can have a huge impact.
arxiv.org/abs/2510.16175
One of my students is empirically investigating this and, as one would expect, it can have a huge impact.
arxiv.org/abs/2510.16175
The Formalism-Implementation Gap in Reinforcement Learning Research
The last decade has seen an upswing in interest and adoption of reinforcement learning (RL) techniques, in large part due to its demonstrated capabilities at performing certain tasks at "super-human l...
arxiv.org
January 29, 2026 at 10:08 AM
This is something I talk about in my paper, where I suggest being explicit about {\gamma}_train (some methods use multiple gammas during training) and \gamma_eval.
One of my students is empirically investigating this and, as one would expect, it can have a huge impact.
arxiv.org/abs/2510.16175
One of my students is empirically investigating this and, as one would expect, it can have a huge impact.
arxiv.org/abs/2510.16175
Reposted by Antonin Raffin
December in Servo…
🎤🧑🏫 FOSDEM talks next week!
🤹🪟 multiple windows
🪆🌐 HTTP proxy support
🔐🕵️ more SubtleCrypto algorithms
💽🗃️ new site data & network API
servo.org/blog/2026/01...
🎤🧑🏫 FOSDEM talks next week!
🤹🪟 multiple windows
🪆🌐 HTTP proxy support
🔐🕵️ more SubtleCrypto algorithms
💽🗃️ new site data & network API
servo.org/blog/2026/01...
January 23, 2026 at 6:39 AM
December in Servo…
🎤🧑🏫 FOSDEM talks next week!
🤹🪟 multiple windows
🪆🌐 HTTP proxy support
🔐🕵️ more SubtleCrypto algorithms
💽🗃️ new site data & network API
servo.org/blog/2026/01...
🎤🧑🏫 FOSDEM talks next week!
🤹🪟 multiple windows
🪆🌐 HTTP proxy support
🔐🕵️ more SubtleCrypto algorithms
💽🗃️ new site data & network API
servo.org/blog/2026/01...
Reposted by Antonin Raffin
Dr. Who plays with Docker How :
docker.how
docker.how
Docker Cheat Sheet — The Ultimate CLI Reference
Comprehensive Docker CLI reference with commands for containers, images, volumes, networks, Compose, and Dockerfile.
docker.how
January 18, 2026 at 8:42 AM
Dr. Who plays with Docker How :
docker.how
docker.how
Reposted by Antonin Raffin
HTML preview & export now available in the web app! With HTML export, you can create a website from the same Typst file as your PDFs. This makes it easy to create documents that feel just as at home on the web as they do in print.
January 13, 2026 at 6:21 PM
HTML preview & export now available in the web app! With HTML export, you can create a website from the same Typst file as your PDFs. This makes it easy to create documents that feel just as at home on the web as they do in print.
Reposted by Antonin Raffin
This network analyzer is very efficient and allows you to find interesting accounts, eg. people followed by lots of the people you follow (but not you).
bsky-follow-finder.theo.io
(Reposting this for folks who have joined Bsky more recently)
bsky-follow-finder.theo.io
(Reposting this for folks who have joined Bsky more recently)
January 12, 2026 at 6:17 PM
This network analyzer is very efficient and allows you to find interesting accounts, eg. people followed by lots of the people you follow (but not you).
bsky-follow-finder.theo.io
(Reposting this for folks who have joined Bsky more recently)
bsky-follow-finder.theo.io
(Reposting this for folks who have joined Bsky more recently)
Reposted by Antonin Raffin
People wanted our Open Source Organizations starter pack to include many projects, so we decided to give them their own starter pack.
go.bsky.app/HvKFRKa
go.bsky.app/HvKFRKa
Open Source projects
Join the conversation
go.bsky.app
January 9, 2026 at 5:00 PM
People wanted our Open Source Organizations starter pack to include many projects, so we decided to give them their own starter pack.
go.bsky.app/HvKFRKa
go.bsky.app/HvKFRKa
"uv is fast because of what it doesn’t do, not because of what language it’s written in"
How did uv get so fast? (Spoiler: not just because it’s written in rust) nesbitt.io/2025/12/26/h...
How uv got so fast
uv’s speed comes from engineering decisions, not just Rust. Static metadata, dropping legacy formats, and standards that didn’t exist five years ago.
nesbitt.io
December 31, 2025 at 4:36 PM
"uv is fast because of what it doesn’t do, not because of what language it’s written in"
Reposted by Antonin Raffin
Using AI coding for data analysis without personal programming skill fills me with dread.
Small errors in the code poisons results in ways that may not be visibly obvious.
LLMs are great when people verify outputs; the path to hell is when they don't.
Small errors in the code poisons results in ways that may not be visibly obvious.
LLMs are great when people verify outputs; the path to hell is when they don't.
December 26, 2025 at 5:07 PM
Using AI coding for data analysis without personal programming skill fills me with dread.
Small errors in the code poisons results in ways that may not be visibly obvious.
LLMs are great when people verify outputs; the path to hell is when they don't.
Small errors in the code poisons results in ways that may not be visibly obvious.
LLMs are great when people verify outputs; the path to hell is when they don't.
Reposted by Antonin Raffin
Hi RL Enthusiasts!
RLC is coming to Montreal, Quebec, in the summer: Aug 16–19, 2026!
Call for Papers is up now:
Abstract: Mar 1 (AOE)
Submission: Mar 5 (AOE)
Excited to see what you’ve been up to - Submit your best work!
rl-conference.cc/callforpaper...
Please share widely!
RLC is coming to Montreal, Quebec, in the summer: Aug 16–19, 2026!
Call for Papers is up now:
Abstract: Mar 1 (AOE)
Submission: Mar 5 (AOE)
Excited to see what you’ve been up to - Submit your best work!
rl-conference.cc/callforpaper...
Please share widely!
RLJ | RLC Call for Papers
rl-conference.cc
December 23, 2025 at 10:16 PM
Hi RL Enthusiasts!
RLC is coming to Montreal, Quebec, in the summer: Aug 16–19, 2026!
Call for Papers is up now:
Abstract: Mar 1 (AOE)
Submission: Mar 5 (AOE)
Excited to see what you’ve been up to - Submit your best work!
rl-conference.cc/callforpaper...
Please share widely!
RLC is coming to Montreal, Quebec, in the summer: Aug 16–19, 2026!
Call for Papers is up now:
Abstract: Mar 1 (AOE)
Submission: Mar 5 (AOE)
Excited to see what you’ve been up to - Submit your best work!
rl-conference.cc/callforpaper...
Please share widely!
Reposted by Antonin Raffin
Almost 5 years in the making... "Hyperparameter Optimization in Machine Learning" is finally out! 📘
We designed this monograph to be self-contained, covering: Grid, Random & Quasi-random search, Bayesian & Multi-fidelity optimization, Gradient-based methods, Meta-learning.
arxiv.org/abs/2410.22854
We designed this monograph to be self-contained, covering: Grid, Random & Quasi-random search, Bayesian & Multi-fidelity optimization, Gradient-based methods, Meta-learning.
arxiv.org/abs/2410.22854
December 17, 2025 at 9:54 AM
Almost 5 years in the making... "Hyperparameter Optimization in Machine Learning" is finally out! 📘
We designed this monograph to be self-contained, covering: Grid, Random & Quasi-random search, Bayesian & Multi-fidelity optimization, Gradient-based methods, Meta-learning.
arxiv.org/abs/2410.22854
We designed this monograph to be self-contained, covering: Grid, Random & Quasi-random search, Bayesian & Multi-fidelity optimization, Gradient-based methods, Meta-learning.
arxiv.org/abs/2410.22854
Reposted by Antonin Raffin
A practical introduction to (deep) RL, providing intuitions to understand the more recent algorithms (continued).
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
RL103: From Deep Q-Learning (DQN) to Soft Actor-Critic (SAC) and Beyond | Antonin Raffin | Homepage
This second blog post continues my practical introduction to (deep) reinforcement learning, presenting the main concepts and providing intuitions to understand the more recent Deep RL algorithms.
In a...
araffin.github.io
December 12, 2025 at 5:47 PM
A practical introduction to (deep) RL, providing intuitions to understand the more recent algorithms (continued).
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
Reposted by Antonin Raffin
What a phenomenal talk by @jenson.org. He works in a very different slice of tech than I do, but his ethos toward developing tech deeply matches my own, and he articulates it so well.
I highly recommend watching it, regardless of whether you're interested in UX.
I highly recommend watching it, regardless of whether you're interested in UX.
My Ubuntu Summit talk is up! Where I talk about:
1. How Desktop UX is effectively dead
2. Why I hate the term UX/UI with the heat of 1000 suns
3. How OSS can actually innovate in #ux
www.youtube.com/watch?v=1fZT...
1. How Desktop UX is effectively dead
2. Why I hate the term UX/UI with the heat of 1000 suns
3. How OSS can actually innovate in #ux
www.youtube.com/watch?v=1fZT...
Are we stuck with the same Desktop UX forever? | Ubuntu Summit 25.10
YouTube video by Canonical Ubuntu
www.youtube.com
December 13, 2025 at 7:58 PM
What a phenomenal talk by @jenson.org. He works in a very different slice of tech than I do, but his ethos toward developing tech deeply matches my own, and he articulates it so well.
I highly recommend watching it, regardless of whether you're interested in UX.
I highly recommend watching it, regardless of whether you're interested in UX.
Reposted by Antonin Raffin
antonin has been cooking olala
A practical introduction to (deep) RL, providing intuitions to understand the more recent algorithms (continued).
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
RL103: From Deep Q-Learning (DQN) to Soft Actor-Critic (SAC) and Beyond | Antonin Raffin | Homepage
This second blog post continues my practical introduction to (deep) reinforcement learning, presenting the main concepts and providing intuitions to understand the more recent Deep RL algorithms.
In a...
araffin.github.io
December 12, 2025 at 6:40 PM
antonin has been cooking olala
A practical introduction to (deep) RL, providing intuitions to understand the more recent algorithms (continued).
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
RL103: From Deep Q-Learning (DQN) to Soft Actor-Critic (SAC) and Beyond | Antonin Raffin | Homepage
This second blog post continues my practical introduction to (deep) reinforcement learning, presenting the main concepts and providing intuitions to understand the more recent Deep RL algorithms.
In a...
araffin.github.io
December 12, 2025 at 5:47 PM
A practical introduction to (deep) RL, providing intuitions to understand the more recent algorithms (continued).
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
In this second post, I continue from DQN on to the Soft Actor-Critic (SAC) algorithm and its extensions.
araffin.github.io/post/rl103/
Reposted by Antonin Raffin
🚀 We just shipped v0.216.0!
Word-level diffing just landed. 🎉
It's been a night-and-day difference for us—seeing exactly what changed within each line.
Word-level diffing just landed. 🎉
It's been a night-and-day difference for us—seeing exactly what changed within each line.
December 10, 2025 at 5:16 PM
🚀 We just shipped v0.216.0!
Word-level diffing just landed. 🎉
It's been a night-and-day difference for us—seeing exactly what changed within each line.
Word-level diffing just landed. 🎉
It's been a night-and-day difference for us—seeing exactly what changed within each line.
An intuitive introduction to flow matching, ICLR 2025 best blog post: iclr-blogposts.github.io/2025/blog/fl...
December 8, 2025 at 6:22 PM
An intuitive introduction to flow matching, ICLR 2025 best blog post: iclr-blogposts.github.io/2025/blog/fl...
New in Stable-Baselines3 documentation: a full example on how to export a trained RL agent to run in the browser, using ONNX and ONNX Runtime Web (JS).
Demo: jonathancoletti.github.io/CarDodgingGym/
Documentation: stable-baselines3.readthedocs.io/en/master/gu...
Demo: jonathancoletti.github.io/CarDodgingGym/
Documentation: stable-baselines3.readthedocs.io/en/master/gu...
December 5, 2025 at 5:39 PM
New in Stable-Baselines3 documentation: a full example on how to export a trained RL agent to run in the browser, using ONNX and ONNX Runtime Web (JS).
Demo: jonathancoletti.github.io/CarDodgingGym/
Documentation: stable-baselines3.readthedocs.io/en/master/gu...
Demo: jonathancoletti.github.io/CarDodgingGym/
Documentation: stable-baselines3.readthedocs.io/en/master/gu...
Reposted by Antonin Raffin
🎤 Announcing the 3rd workshop on Reinforcement Learning in Mannheim 🎤
We have an amazing lineup of speakers: @Mathieugeist, @gio_ramponi, Theresa Eimer, @SarahKeren_, @araffin2, @c_rothkopf, and @AdrienBolland
⏰ Friday 6th February
📍University of Mannheim
We have an amazing lineup of speakers: @Mathieugeist, @gio_ramponi, Theresa Eimer, @SarahKeren_, @araffin2, @c_rothkopf, and @AdrienBolland
⏰ Friday 6th February
📍University of Mannheim
December 2, 2025 at 11:45 AM
🎤 Announcing the 3rd workshop on Reinforcement Learning in Mannheim 🎤
We have an amazing lineup of speakers: @Mathieugeist, @gio_ramponi, Theresa Eimer, @SarahKeren_, @araffin2, @c_rothkopf, and @AdrienBolland
⏰ Friday 6th February
📍University of Mannheim
We have an amazing lineup of speakers: @Mathieugeist, @gio_ramponi, Theresa Eimer, @SarahKeren_, @araffin2, @c_rothkopf, and @AdrienBolland
⏰ Friday 6th February
📍University of Mannheim
Reposted by Antonin Raffin
Exciting workshop for RL enthusiasts in Mannheim! 👇
Workshop on Reinforcement Learning 2026, taking place on 𝐅𝐞𝐛𝐫𝐮𝐚𝐫𝐲 𝟔, 𝟐𝟎𝟐𝟔, at the 𝐔𝐧𝐢𝐯𝐞𝐫𝐬𝐢𝐭𝐲 𝐨𝐟 𝐌𝐚𝐧𝐧𝐡𝐞𝐢𝐦, Germany.
Participation in the workshop is 𝐟𝐫𝐞𝐞 𝐨𝐟 𝐜𝐡𝐚𝐫𝐠𝐞!
Check the program and register: www.wim.uni-mannheim.de/doering/conf...
Workshop on Reinforcement Learning 2026, taking place on 𝐅𝐞𝐛𝐫𝐮𝐚𝐫𝐲 𝟔, 𝟐𝟎𝟐𝟔, at the 𝐔𝐧𝐢𝐯𝐞𝐫𝐬𝐢𝐭𝐲 𝐨𝐟 𝐌𝐚𝐧𝐧𝐡𝐞𝐢𝐦, Germany.
Participation in the workshop is 𝐟𝐫𝐞𝐞 𝐨𝐟 𝐜𝐡𝐚𝐫𝐠𝐞!
Check the program and register: www.wim.uni-mannheim.de/doering/conf...
November 25, 2025 at 1:51 PM
Exciting workshop for RL enthusiasts in Mannheim! 👇
Workshop on Reinforcement Learning 2026, taking place on 𝐅𝐞𝐛𝐫𝐮𝐚𝐫𝐲 𝟔, 𝟐𝟎𝟐𝟔, at the 𝐔𝐧𝐢𝐯𝐞𝐫𝐬𝐢𝐭𝐲 𝐨𝐟 𝐌𝐚𝐧𝐧𝐡𝐞𝐢𝐦, Germany.
Participation in the workshop is 𝐟𝐫𝐞𝐞 𝐨𝐟 𝐜𝐡𝐚𝐫𝐠𝐞!
Check the program and register: www.wim.uni-mannheim.de/doering/conf...
Workshop on Reinforcement Learning 2026, taking place on 𝐅𝐞𝐛𝐫𝐮𝐚𝐫𝐲 𝟔, 𝟐𝟎𝟐𝟔, at the 𝐔𝐧𝐢𝐯𝐞𝐫𝐬𝐢𝐭𝐲 𝐨𝐟 𝐌𝐚𝐧𝐧𝐡𝐞𝐢𝐦, Germany.
Participation in the workshop is 𝐟𝐫𝐞𝐞 𝐨𝐟 𝐜𝐡𝐚𝐫𝐠𝐞!
Check the program and register: www.wim.uni-mannheim.de/doering/conf...
Reposted by Antonin Raffin
TMLR (@tmlrorg.bsky.social) is now proud to support interactive HTML-based submissions, going "Beyond PDF" -- check it out!
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
🚀 Introducing TMLR Beyond PDF!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
🎬 This is a new, HTML-based submission format for TMLR, that supports interactive figures and videos, along with the usual LaTeX and images.
🎉 Thanks to TMLR Editors in Chief: Hugo Larochelle, @gautamkamath.com, Naila Murray, Nihar B. Shah, and Laurent Charlin!
November 25, 2025 at 4:14 PM
TMLR (@tmlrorg.bsky.social) is now proud to support interactive HTML-based submissions, going "Beyond PDF" -- check it out!
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.
Thanks to Paul Vicol (@paulvicol.bsky.social) for his tireless work on this new option, as well as the OpenReview team.