✨ The last day kicked off with an amazing talk by @katjahofmann.bsky.social "World and Human Action Models for Gameplay Ideation" 🎮🤖 Exciting vision from the Game Intelligence team @msftresearch.bsky.social

I am in Vancouver at ICML, and tomorrow I will present our newest paper "Partially Observable Reinforcement Learning with Memory Traces". We argue that eligibility traces are more effective than sliding windows as a memory mechanism for RL in POMDPs. 🧵

After all these reports of authors adding language instructions for LLM reviews in their papers I wanted to check this myself and I downloaded the .tex source from one of these papers. Here is an example. (I will not share the identity of the paper)

WHAT ELSE DID THEY LIE ABOUT - dashare.zone ADMIN

Join us for Nneka's presentation tomorrow! Last talk before the summer break.

new preprint with the amazing @lviano.bsky.social and @neu-rips.bsky.social on offline imitation learning! learned a lot :) when the expert is hard to represent but the environment is simple, estimating a Q-value rather than the expert directly may be beneficial. lots of open questions left though!

new work on computing distances between stochastic processes ***based on sample paths only***! we can now: - learn distances between Markov chains - extract "encoder-decoder" pairs for representation learning - with sample- and computational-complexity guarantees read on for some quick details.. 1/n

"The chatbot responded that it was perfectly okay for landlords to discriminate based on whether those potential tenants need rental assistance" From "Government officials are letting AI do their jobs. Badly," by @emilymbender.bsky.social & @alexhanna.bsky.social thebulletin.org/2025/05/gove...

Disabled people have been pleading with folks to show solidarity for years. We saw the rising tide of eugenics the moment everyone declared Covid over & decided to leave us by the wayside. Pandemics give rise to fascism. What we’re seeing is the result of allowing disabled folks to be left behind

found a cool game theory book arxiv.org/abs/1512.06808

Trying to get my head round how this could make it to print. Many warning signs for all of us

In the year of our Lord 2025... we are still putting on makeup on women without their consent. It's truly like feminism never happened 🙅‍♀️ (I have been ranting about this ever since people started using it to demonstrate GANs back in the 2010s, I can't believe it's still an "acceptable" task in AI!)

I have started a substack lately on my random thoughts, if anyone's interested <3

Singapore book haul! (+ St Jordi)

Had a lot of fun presenting our work at "Offline RL in Regular Decision Processes: Sample Efficiency via Language Metrics" at #ICLR2025, with my co-authors Alessandro Ronca and Sadegh Talebi! check out our paper here: openreview.net/forum?id=EW6...

"It's wonderful watching the well of knowledge being poisoned in real time". 📷 Emily Gorcenski

First draft online version of The RLHF Book is DONE. Recently I've been creating the advanced discussion chapters on everything from Constitutional AI to evaluation and character training, but I also sneak in consistent improvements to the RL specific chapter. rlhfbook.com

A mid-week self reminder

A relatedly bad idea is working on research you don't deeply believe in for pragmatic reasons. You start with "oh, it'll be a quick low-hanging fruit project" and 9 months later you're still working on it and in despair

Took a weekend off to travel to Bruges. It was beautiful :O

Mark your calendars, EWRL is coming to Tübingen! 📅 When? September 17-19, 2025. More news to come soon, stay tuned!

Thought-canceling headphones

Beautiful.

Congrats to this year's Turing award winners! www.nytimes.com/2025/03/05/t... Incidentally, if you'd like to hear from them, we know a place they've given / are giving keynotes

Made a small drawing in memory of my partner’s cat :’)

does anyone have a poster template they absolutely swear by? pls link

I wish I could tell the moon how much I like it

Happy PLDI author response night to those who celebrate

Heard a song from the 80s reference a "long distance call" and it occurred to me that this is a concept that no longer holds any emotional currency whatsoever

Today (21st Feb) is the International Mother Language day, commemorating students of Bangladesh, who were brutally murdered by the police in their fight to preserve their mother tongue Bangla. (check Alt text for translation of the poster!)

Going through my paper to correct a notation feels awfully similar to Macro Data Refinement. #thisfeelsoff

Every podcast I don’t listen to is called “Achieving the Limit with Duff and Ween” and gets sold for 300 million dollars while every podcast I do listen to is called “The Movie Toilet” and begs me to buy a t shirt so they can eat food

for me it was when i realized a lot of the movies i loved as a kid were propaganda for military recruitment

Model-free deep RL algorithms like NFSP, PSRO, ESCHER, & R-NaD are tailor-made for games with hidden information (e.g. poker). We performed the largest-ever comparison of these algorithms. We find that they do not outperform generic policy gradient methods, such as PPO. arxiv.org/abs/2502.08938 1/N

this is such a good piece

any international/intergovernmental report on “AI safety” that fails to acknowledge/ground recommendation on the growing unprecedented corporate power and monopoly is not a friend of the public, civil society, and definitely of marginalised and impacted groups

“My death as a member of the third gender and transgender population does not mean you won. It solely marks the end of our association. You are not the land of the free.” zerodoesntsleep.substack.com/p/vet-found-...

I think I've found a way to articulate what bothers me about the trajectory of the modern tech industry (and to some extent... society). We are on this Earth to provide for each other. That's the "mission" of humanity as I see it. This is incompatible with what the tech industry seeks to do.

Ok all I want from @bsky.app is an algorithmic feed that allows me to explicitly filter out US political news. Science communities are only going to tolerate feeds dominated by funding outrages for so long (or we will keep doomscrolling forever)

Today I discovered argmin.net and I’m gonna read everything on it www.argmin.net/p/going-beyo...

The rise of ChatGPT isn't just about AI—it's about how cluttered the web has become. Compare these recipes: ChatGPT gives me ingredients instantly, while a cooking website shows me an auto-playing video and a prompt to log into Google so I can see an ad

The debugging progression of "this has been working fine for months" --> "suddenly things do not work" --> "how did this ever work" is Hegelian dialectic.

Love the AI image labelling by @aimod.social. I prefer not interacting with accounts using AI images as profile picture/banner.

Most of the world’s brightest minds are hard at work developing new ways for you to get depression from a computer

We've built a simulated driving agent that we trained on 1.6 billion km of driving with no human data. It is SOTA on every planning benchmark we tried. In self-play, it goes 20 years between collisions.

PROBLY FINE - dashare.zone ADMIN

Truly life-changing labels to hide/warn about any post related to US politics. Am at peace. @uspol.bluesky.bot

RLC call for workshops is out rl-conference.cc/callforworks...! Submissions open on Feb 3rd, deadline on March 7, with the workshops on Aug. 5th. Last year's workshops were inimitable and we @claireve.bsky.social and Josiah Hanna) look forward to your amazing proposals

As an immigrant studying in a country where English is not the first language, it’s important for the host university to make these resources accessible and affordable for all

If you’re in grad school, finding a therapist can be really helpful. The thing you’re doing is hard and it’s harder if you don’t have help managing imposter syndrome, stress, self esteem, and a whole bunch of other things.