@michaelddennis.bsky.social
RS DeepMind. Works on Unsupervised Environment Design, Problem Specification, Game/Decision Theory, RL, AIS. prev CHAI_Berkeley
Pinned
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
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 future is agents in generative environments
Reposted
*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!
🚨 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!
July 8, 2025 at 5:16 PM
*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!
🚨 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!
Reposted
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
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
Oskar Stålberg: Landscapes of Hex and Square - Summer School on AI and Games 2023
YouTube video by Summer School on AI and Games TV
www.youtube.com
February 26, 2025 at 6:09 PM
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
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
Reposted
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...
arxiv.org/abs/2502.096...
February 17, 2025 at 6:26 PM
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...
arxiv.org/abs/2502.096...
Reposted
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!
A theory of appropriateness with applications to generative artificial intelligence
arxiv.org/abs/2412.19010
And happy new year everyone!
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...
arxiv.org
December 31, 2024 at 7:48 AM
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!
A theory of appropriateness with applications to generative artificial intelligence
arxiv.org/abs/2412.19010
And happy new year everyone!
Reposted
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! 🧞
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
December 4, 2024 at 4:13 PM
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! 🧞
Reposted
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!
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!
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
December 4, 2024 at 4:07 PM
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!
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 future is agents in generative environments
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
December 4, 2024 at 5:53 PM
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 future is agents in generative environments
Reposted
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
November 19, 2024 at 7:05 PM
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
Reposted
Welcome! @natashajaques.bsky.social. 👋
I think you may find one or two people who share your sentiment here... 😁
I think you may find one or two people who share your sentiment here... 😁
One of my first posts on twitter was "fuck twitter". I'd just like to reiterate that sentiment today, as I join bluesky
November 19, 2024 at 1:07 AM
Welcome! @natashajaques.bsky.social. 👋
I think you may find one or two people who share your sentiment here... 😁
I think you may find one or two people who share your sentiment here... 😁