Sobhan Mohammadpour

[NYU Center for Data Science @NYUDataScience]

Simple deep RL was thought to fail in imperfect-info games like poker.

A new ICLR 2026 paper shows that with proper tuning, generic methods like PPO match or beat specialized approaches like fictitious play and counterfactual regret minimization.

https://t.co/RvkZ6XxR6J

1/2

[Co-op Tory 🍁 @CoopTory]

It is becoming increasingly apparent that Carney is a modern Castlereagh.

[Sobhan Mohammadpour @___sobhan___]

@msharmavikram It was really nice that the guide was a single html page.

[Sobhan Mohammadpour @___sobhan___]

Never a dull moment

[Sobhan Mohammadpour @___sobhan___]

Another day, another

[Alexey Zaytsev @xl0xl0xl0]

TIL, this is actually valid Python.

[Chad Scherrer @ChadScherrer]

UNICODE NEEDS BETTER SUBSCRIPT/SUPERSCRIPT COVERAGE

https://t.co/TuE31994Fp

Easy ways to help make this happen:
• Retweet this
• Star the repo
• Show it to others

[Sobhan Mohammadpour @___sobhan___]

@qubitium @BayesWatch & co had a blog post on the Julia blog about something somewhat similar https://t.co/YmYKRpEWWf They observed between 30x to 5x improvements