*New Paper*

🚨 Goal misgeneralization occurs when AI agents learn the wrong reward function, instead of the human's intended goal.

😇 We show that training with a minimax regret objective provably mitigates it, promoting safer and better-aligned RL policies!

YouTube video by Summer School on AI and Games TV Oskar Stålberg: Landscapes of Hex and Square - Summer School on AI and Games 2023

a very interesting multi-criteria optimisation problem for procedural level generation in 3D games

could one be inspiration for the environment design as a general sum game or a coalition game, going beyond zero-sum stuff?

www.youtube.com/watch?v=Npfo...

cc @michaelddennis.bsky.social

In our recent paper, we outline 15 fully risk-agnostic, process-based, evidence-seeking policy objectives. None of them limit *what* developers can do -- they just affect *reporting* and *visibility.*
arxiv.org/abs/2502.096...

A theory of appropriateness with applications to generative artificial intelligence What is appropriateness? Humans navigate a multi-scale mosaic of interlocking notions of what is appropriate for different situations. We act one way with our friends, another with our family, and yet...

Very happy to announce the publication of our latest paper:

A theory of appropriateness with applications to generative artificial intelligence
arxiv.org/abs/2412.19010

And happy new year everyone!

Not at NeurIPS but would love to chat about it sometime! sounds right, and I think a lot of weird stuff happens when you take it to the extreme

Excited to reveal Genie 2, our most capable foundation world model that, given a single prompt image, can generate an endless variety of action-controllable, playable 3D worlds. Fantastic cross-team effort by the Open-Endedness Team and many other teams at Google DeepMind! 🧞

Thrilled to share Genie 2! Endless environments created by text or images, a key to open-ended/AI-Generating Algorithms.

Genie 1 showed it's possible. 9 months later, Genie 2 shows jaw-dropping progress.🤯 Witness the magic of scale, again. 📈🚀 Thx to all team members @deep-mind.bsky.social!

It’s been a crazy 2 years seeing so many amazingly talented researchers bring GENerative Interactive Environments alive in Genie 1 and 2.

the future is agents in generative environments

The secret to doing good research is always to be a little underemployed. You waste years by not being able to waste hours. - Amos Tversky

Welcome! @natashajaques.bsky.social. 👋

I think you may find one or two people who share your sentiment here... 😁

🤣 really appreciate the support ❤️

Thanks Eugene! I just have to get around to starting a blog already to post my weirdest takes 😅

just joined bluesky too! 🙂

Always love a classic “more environments” when they admit the current envs prove the method works

I have seen the silliest. strengths, weaknesses, and questions all a variant on the same sentence