The Automated RL Workshop Coming to RLC'26!
- Tweets 54
- Following 15
- Followers 165
- Likes 23
AutoRL Workshop retweeted
Thanks to everyone who joined @AutoRL_Workshop, it was really great to see the room full and everyone actively participating. Until the next one!
Really enjoyed working with Julian on the workshop, and I’m looking forward to the next chapters ahead 🥳
Excited for Day 2 of #RLC! 🚀
To conclude an amazing day of talks, we have our final speaker, @araffin2, sharing his perspective on “Making RL Work Out-of-the-Box.” 🚀
A fitting way to wrap up the workshop: looking at what it actually takes to make RL work reliably in practice , because RL is hard! 😄
#RLC2026
4
36
5,859
AutoRL Workshop retweeted
To conclude an amazing day of talks, we have our final speaker, @araffin2, sharing his perspective on “Making RL Work Out-of-the-Box.” 🚀
A fitting way to wrap up the workshop: looking at what it actually takes to make RL work reliably in practice , because RL is hard! 😄
#RLC2026
We’re back from the break with @junh_oh , speaking about “Discovering State-of-the-Art Reinforcement Learning Algorithms”, recently published in Nature. 🚀
Really cool work on framing algorithm discovery itself as an RL problem, using meta-learning to discover learning rules rather than designing them by hand.
2
8
5,790
AutoRL Workshop retweeted
We’re back from the break with @junh_oh , speaking about “Discovering State-of-the-Art Reinforcement Learning Algorithms”, recently published in Nature. 🚀
Really cool work on framing algorithm discovery itself as an RL problem, using meta-learning to discover learning rules rather than designing them by hand.
Lats talk before the lunch break! @Theo_Vincent_ is leading a great discussion on automating RL algorithm design: Can adaptive methods like AdaDQN outperform fixed RL algorithms and keep adapting in continual learning settings? ⚡️
3
13
1,738
AutoRL Workshop retweeted
Starting now!! 🥳
🔥 AutoRL Debate, not your usual panel!
We’re switching things up this afternoon. Instead of a traditional panel, our six participants will debate three controversial positions in AutoRL, with a twist: they’ll be randomly assigned to argue for or against each one, regardless of their own views. 👀
Expect opposing arguments, lively discussion, and hopefully a few surprising insights along the way.
Come join us!! It is happening at 2:00 PM (Montreal time).
Led by our amazing @pcastr! 🔥
2
19
2,357
AutoRL Workshop retweeted
Come by, starting in 20ish minutes! I'm currently making my way from the airport but should arrive just in the nick of time 😅
Looking forward to moderating this debate!
🔥 AutoRL Debate, not your usual panel!
We’re switching things up this afternoon. Instead of a traditional panel, our six participants will debate three controversial positions in AutoRL, with a twist: they’ll be randomly assigned to argue for or against each one, regardless of their own views. 👀
Expect opposing arguments, lively discussion, and hopefully a few surprising insights along the way.
Come join us!! It is happening at 2:00 PM (Montreal time).
Led by our amazing @pcastr! 🔥
1
12
2,260
AutoRL Workshop retweeted
🔥 AutoRL Debate, not your usual panel!
We’re switching things up this afternoon. Instead of a traditional panel, our six participants will debate three controversial positions in AutoRL, with a twist: they’ll be randomly assigned to argue for or against each one, regardless of their own views. 👀
Expect opposing arguments, lively discussion, and hopefully a few surprising insights along the way.
Come join us!! It is happening at 2:00 PM (Montreal time).
Led by our amazing @pcastr! 🔥
3
17
5,923
AutoRL Workshop retweeted
Lats talk before the lunch break! @Theo_Vincent_ is leading a great discussion on automating RL algorithm design: Can adaptive methods like AdaDQN outperform fixed RL algorithms and keep adapting in continual learning settings? ⚡️
Next up, we have @clarelyle giving a talk on “Optimization and Its Discontents,” taking a closer look at a fundamental question in RL: What should we reinforce, and how should we reinforce it?
Lots to rethink about the optimization choices we often take for granted in RL. 🤔
#RLC2026
3
14
2,273
AutoRL Workshop retweeted
Starting now with our first speaker at the @AutoRL_Workshop ! 🚀
@white_martha kicks us off with “Reconsidering Hyperparameter Tuning for Deployment on Physical Systems”, exploring what it takes to bring RL beyond benchmarks and into real-world physical systems.
#RLC2026
Excited to be at #RLC2026! 🎉
Day 1 is kicking off with so many great workshops and conversations. Looking forward to catching up with familiar faces, meeting new ones, and chatting about all things RL!
If you’re around, come say hi 👋 , and don’t miss @RLFrameWorkshop and @AutoRL_Workshop. Hope to see you there!
2
14
3,840
AutoRL Workshop retweeted
Next up, we have @clarelyle giving a talk on “Optimization and Its Discontents,” taking a closer look at a fundamental question in RL: What should we reinforce, and how should we reinforce it?
Lots to rethink about the optimization choices we often take for granted in RL. 🤔
#RLC2026
Starting now with our first speaker at the @AutoRL_Workshop ! 🚀
@white_martha kicks us off with “Reconsidering Hyperparameter Tuning for Deployment on Physical Systems”, exploring what it takes to bring RL beyond benchmarks and into real-world physical systems.
#RLC2026
1
2
24
4,100
Finally, here are the participants of our AutoRL debate!
We’re switching things up from a traditional panel: three controversial positions, with participants randomly assigned to argue for or against each one.
Expect opposing views, lively discussions, and surprising insights!
1
4
21
9,032
Meet our debate participants: @white_martha @clarelyle @Theo_Vincent_ @junh_oh @smdvln and @mcbeukman.
The debate will be moderated by @pcastr.
6
211
Our AutoRL workshop is happening this weekend!
Here’s the final lineup of speakers: @white_martha, @Theo_Vincent_, @clarelyle, @junh_oh and @araffin2 will share their expertise on real-world applications, meta-learning, robustness, and much more.
Don’t miss it! See you there!
5
21
2,688
RLC in Montreal is right around the corner! 🍁
To kick off our own countdown, we’ll be introducing our invited speakers and panelists throughout the week.
And for the panel, we’ve come up with an exciting format that we can’t wait to share! 🚀
1
10
668
We extend the submission deadline to May 29, 2026 (AoE), given the number of requests. We look forward to your submissions!
5
1,286
Reminder: our submission deadline is approaching fast — 22 May (AOE)!
Submit your work on making RL truly out-of-the-box. We are particularly interested in research at the intersection of RL and LLMs to improve robustness and adaptability.
sites.google.com/view/automa…
1
7
319
What even is AutoRL and what kind of submissions are we looking for? 🤔
In theory, it should be easy to use a successful algorithm like PPO that can solve many RL tasks on a new problem, but this is rarely the case. Instead, a lot of effort and expertise is needed each time.
1
1
7
435
AutoRL methods try to close that gap in various ways: meta-learning, stabilization techniques, optimizing hyperparameters, or gathering new insights about how reproducibility works in RL.
What's important is not the method, but the goal: making RL an out-of-the-box technology!
1
5
124
If that's something you're passionate about as well, consider submitting to the workshop!
The submission deadline is the 22nd of May, the workshop takes place on the 15th of August.
You can find our submission instructions here: sites.google.com/view/automa…
5
109
🔥 AutoRL Workshop returns to RLC 2026 in Montréal 🇨🇦
Join us to tackle RL brittleness and advance methods that work “out of the box”.
More info: sites.google.com/view/automa…
This year's organisers are: Theresa Eimer, @DierkesJul67648, @johanobandoc, @pcastr, @HolgerHoos
1
10
20
7,845
We have a speaker change: instead of @jparkerholder we'll hear from @MichaelD1729 - the focus of the talk is the same, though, so join if you're interested in generating any environments you can imagine!
4
14
1,984