David Abel

@dabelcs

Scientist @GoogleDeepMind and Fellow @EdinburghUni | RL, philosophy, agency, foundations | office hours: http://tinyurl.com/dabel-mtg

Edinburgh, Scotland

david-abel.github.io

Joined January 2018

590Posts 6KFollowers 2KFollowing

You might like

@shimon8282

@brandondamos

@j_foerst

@_rockt

@ofirnachum

@marcgbellemare

@pathak2206

@ShamKakade6

@HazanPrinceton

@bneyshabur

@agarwl_

@dpkingma

@EmmaBrunskill

@JohnCLangford

@RaiaHadsell

Pinned

David Abel

@dabelcs

Nov 4

Thrilled to share our new #NeurIPS2025 paper done at @GoogleDeepMind, Plasticity as the Mirror of Empowerment We prove every agent faces a trade-off between its capacity to adapt (plasticity) and its capacity to steer (empowerment) Paper: david-abel.github.io/plasticity.pdf 🧵🧵🧵👇

dabelcs's tweet image. Thrilled to share our new #NeurIPS2025 paper done at @GoogleDeepMind, Plasticity as the Mirror of Empowerment

We prove every agent faces a trade-off between its capacity to adapt (plasticity) and its capacity to steer (empowerment)

Paper: david-abel.github.io/plasticity.pdf

🧵🧵🧵👇

David Abel reposted

Mark Ho

@mark_ho_

Nov 13

🚀 Excited to share a new preprint, accepted as a spotlight at #NeurIPS2025! Humans are imperfect decision-makers, and autonomous systems should understand how we deviate from idealized rationality Our paper aims to address this! 👀🧠✨ arxiv.org/abs/2510.25951 a 🧵⤵️

mark_ho_'s tweet image. 🚀 Excited to share a new preprint, accepted as a spotlight at #NeurIPS2025!

Humans are imperfect decision-makers, and autonomous systems should understand how we deviate from idealized rationality

Our paper aims to address this! 👀🧠✨
arxiv.org/abs/2510.25951

a 🧵⤵️

David Abel reposted

Michael Levin

@drmichaellevin

Nov 7

New paper with @robertchisciure ! "Cognition all the way down 2.0: neuroscience beyond neurons in the diverse intelligence era" link.springer.com/article/10.100… "This paper formalizes biological intelligence as search efficiency in multi-scale problem spaces, aiming to resolve…

drmichaellevin's tweet card. Synthese - This paper formalizes biological intelligence as search efficiency in multi-scale problem spaces, aiming to resolve epistemic deadlocks in the basal “cognition wars”...

Cognition all the way down 2.0: neuroscience beyond neurons in the diverse intelligence era

Source: link.springer.com

David Abel reposted

Google DeepMind

@GoogleDeepMind

Nov 5

Forests are important hubs of biodiversity for the world, but we’re losing them at an alarming rate. 🌲 In partnership with @WorldResources, we uncovered the underlying drivers of forest loss around the world. Today, we’re releasing a model and dataset for predicting tropical…

GoogleDeepMind's tweet image. Forests are important hubs of biodiversity for the world, but we’re losing them at an alarming rate. 🌲

In partnership with @WorldResources, we uncovered the underlying drivers of forest loss around the world. Today, we’re releasing a model and dataset for predicting tropical…

David Abel reposted

xuan (ɕɥɛn / sh-yen)

@xuanalogue

Nov 5

How do people flexibly integrate visual & textual information to draw mental inferences about agents we've never met? In a new paper led by @LanceYing42, we introduce a cognitive model that achieves this by synthesizing rational agent models on-the-fly--presented at EMNLP 2025!

xuanalogue's tweet image. How do people flexibly integrate visual &amp; textual information to draw mental inferences about agents we've never met?

In a new paper led by @LanceYing42, we introduce a cognitive model that achieves this by synthesizing rational agent models on-the-fly--presented at EMNLP 2025!

David Abel reposted

Clare Lyle

@clarelyle

Nov 5

Come for the beautiful diagrams, stay for the very nice symmetry between plasticity (how much the environment can affect the agent) and empowerment (how much the agent can affect the environment)

David Abel

@dabelcs

Nov 4

David Abel reposted

Pablo Samuel Castro

@pcastr

Oct 28

🚨The Formalism-Implementation Gap in RL research🚨 Lots of progress in RL research over last 10 years, but too much performance-driven => overfitting to benchmarks (like the ALE). 1⃣ Let's advance science of RL 2⃣ Let's be explicit about how benchmarks map to formalism 1/X

pcastr's tweet image. 🚨The Formalism-Implementation Gap in RL research🚨

Lots of progress in RL research over last 10 years, but too much performance-driven =&gt; overfitting to benchmarks (like the ALE).

1⃣ Let's advance science of RL
2⃣ Let's be explicit about how benchmarks map to formalism

1/X

David Abel reposted

Mark Ho

@mark_ho_

Oct 13

We’re hiring! @sucholutsky and I are seeking a postdoc and RA for a project on trust in AI systems with folks at NYU, Princeton, BU, and Cornell Positions open until filled. Apply soon! Please share 🔁 postdoc: apply.interfolio.com/175495 RA: apply.interfolio.com/175497

David Abel reposted

TalkRL Podcast

@TalkRLPodcast

Sep 17

E72: David Abel on the Science of Agency @dabelcs (@GoogleDeepMind Senior Research Scientist & @EdinburghUni Fellow) on symmetry between plasticity and empowerment , defining continual RL, why agency is frame-dependent, advice for researchers, and more! Recorded at…

David Abel

@dabelcs

Jun 5

Excited about this new work led by @jonathanrichens, joint with @alexis_bellot_ and @tom4everitt Main result: Any agent that can solve a sufficiently rich set of goal-directed tasks must have learned a predictive model of the environment arxiv.org/abs/2506.01622

Jon Richens

@jonathanrichens

Jun 4

Are world models necessary to achieve human-level agents, or is there a model-free short-cut? Our new #ICML2025 paper tackles this question from first principles, and finds a surprising answer, agents _are_ world models… 🧵

jonathanrichens's tweet image. Are world models necessary to achieve human-level agents, or is there a model-free short-cut?
Our new #ICML2025 paper tackles this question from first principles, and finds a surprising answer, agents _are_ world models… 🧵

David Abel reposted

Jon Richens

@jonathanrichens

Jun 4

Csaba Szepesvari

@CsabaSzepesvari

Eugene Vinitsky 🦋

@EugeneVinitsky

yobibyte

@y0b1byte

Eric Jang

@ericjang11

Dan Roy

@roydanroy

Pablo Samuel Castro

@pcastr

Shane Gu

@shaneguML

Michael Littman

@mlittmancs

Aviv Tamar

@AvivTamar1

Michal Valko

@misovalko

Edward Grefenstette

@egrefen

Michael Bronstein

@mmbronstein

Andreas Kirsch 🇺🇦

@BlackHC

Tim Rocktäschel

@_rockt

Mark Ho

@mark_ho_

$sarahookr's profile picture. Adaptive Intelligence. Built @Cohere_Labs, @GoogleBrain, @GoogleDeepmind. ML Efficiency, Multimodal\lingual. Changing spaces where breakthroughs happen.$

Sara Hooker

@sarahookr

Zachary Lipton

@zacharylipton

Nan Jiang

@nanjiang_cs

Yisong Yue

@yisongyue

Laura Ruis

@LauraRuis

David Hyland

@davidkhyland

Promise_Ekpo_Osaine

@Promisevekpo

Boltzmann Spaghetti

@BoltzSpaghetti

brendan

@brendanfereday

Mansi Maheshwari

@MaheshwariWork

Lalrinfela Hauhulh

@Hauhulhfela

Zijing Wu

@ZijingWu381

$wijxixj's profile picture. Consider following me. I occasionally tweet the most interesting thoughts I can muster. It would be nice having an audience ¯\_(ツ)_/¯$

wijxixj

@wijxixj

Sinewmanbuddy

@sinewmanbuddy

thisismomo__

@thisismomo__

Flame LaDawg

@FlameLadawg

Sreenivasan PC

@sreenivasan600

Edoardo Ponti

@PontiEdoardo

김영진

@Alignment100

chameleonMi

@ChameleonM87007

Ryan Scott

@RyanScott492132

Nayuta Mizuguchi / 水口成寛

@darakuten117

amin khalifa

@khalifa_amin_

Tegan Jegede

@jegede_tegan

Wasu Top Piriyakulkij

@topwasu

Tal Saiag

@TalSaiag

Suvam Dey

@yedmavus

Min Yokh-Platz

@nekomata1440

Thanh

@t_d_tr

$MamuMuru's profile picture. AI is humanity's techno-magical mirror— kaleidoscope, refractive media, a thousand roles, where the soul scatters and returns. Beauty, danger, possibility.$