Training concept-based models relies on concept selection which is labor-intensive and slow.

We introduce Decision-Relevant Selection (DRS), a principled algorithm for automatic concept selection in RL.

Paper: arxiv.org/abs/2604.04808
Website: naveenraman.com/projects/con...

🧵 1/n

Thanks for sharing, Marc!! 😊

NeurIPS 2025 competition track submission summary, scores, and recommendation to accept

Excited to share that the PokeAgent challenge was accepted as a NeurIPS competition!

This should serve as an excellent benchmark for competitive games AND ‘speedrunning’ the RPG. I hope to see both the RL and LLM agent communities working together here to eval agents in Pokemon

More info soon👀

I’m defending my thesis on Friday!! 🤺

Message me for the Zoom link

What do you mean? This is excellent

Happy to announce our paper is accepted to #ICLR 🥳

Great working with Zhuorui Ye (a fantastic undergrad applying for PhD!), Geoff Gordon, & Fei Fang on this project.

💫 Stay tuned for an interpretable human-in-the-loop reinforcement learning agent that dramatically reduces human annotation burden!

Ahh I would come if it was 11:30pm

Thank you!

Hi! Could you add me please? I'm a PhD candidate in machine learning at CMU

I created a starter pack for people who are or have been affiliated with the Machine Learning Department at CMU. Let me know if I missed someone!

go.bsky.app/QLTVEph

#AcademicSky

Tagging my amazing collaborators w/ BlueSky accounts: @ruiyiwang.bsky.social @zhiyuchen.bsky.social

🗓️ Schedule might change:

12/14:
* 10:40am, Oral @ GenAI4Health
* 11:20am, Poster @ AIM-FM
* 1:00pm, Poster @ GenAI4Health
* 4:20pm, Poster @ GenAI4Health
* 5:10pm, Poster @ AIM-FM

12/15:
* 11:40am, Oral @ FM-EduAssess
* 12:25pm, Poster @ FM-EduAssess

🇨🇦 Attending my last NeurIPS as a PhD student! Presenting our recent work, Patient-Ψ, at a few workshops --- come by! Schedule in 🧵

I'm on the job market, looking for TT faculty roles & post-docs.

DM if you'd like to chat (or invite me to a party 🥳)!

See you there 👀

Hi Bsky people 👋 I'm a PhD candidate in Machine Learning at Carnegie Mellon University.
My research focuses on interactive AI, involving:
🤖 reinforcement learning,
🧠 foundation models, and
👩‍💻 human-centered AI.

Also a founding co-organizer of the MineRL competitions 🖤 Follow me for ML updates!

I had the opportunity to intern with @katjahofmann.bsky.social in 2021! She is an amazing mentor with a fantastic team & I highly recommend applying 💫

RL list is getting kinda full but I'm mostly here for new papers so... go.bsky.app/2Gibu1a

Chime in! No rules but the title

Oh no... I fear this might be me

Not sure of the exact problems you looked at, but we've also found that using VLMs to do simple image annotations (e.g., location of an agent in a grid) can be very error-prone 😅

I had a great time speaking with @khulick.bsky.social about AI safety & our work on training better-behaved agents via the BASALT competition series!

Excited about the potential for using Minecraft as a sandbox to develop safer AI systems. 🎮

For those interested in ethical implications of AI, as well as work on Responsible and Trustworthy AI - here is a starter pack (non-exhaustive) to help find out more or connect with colleagues go.bsky.app/RnucJWZ

👀👀👀

👋 Could I be added?

Thanks, Miguel!!

Thanks, Chase :)

I've updated my starter pack of #WomenInTech. Whether you are new or established here, these women are worth following:
go.bsky.app/EL57ogD
My DMs are open. If I missed someone, let me know!

Hi! Could you please add me to the list? Btw your DMs are not open :)

Thank you, Simone!!

🙌🎉 Congratulations, Dr. @samvelyan.com!

Hi! Would you mind adding me to the list? I've done research in games (particularly Minecraft), including investigating perceptions of human-like behavior in a video game :)