Max Taylor-Davies
@MaxTaylorDavie1
Followers
218
Following
4K
Media
12
Statuses
194
phd student @infAtEd working on social {learning, cognition} // prev @imperialcollege, @ucl // jazz enjoyer
🏴
Joined July 2021
3 papers from @FlowersINRIA are presented this week at the @ALifeConf , leveraging curiosity-driven AI to explore complex behaviors in #Lenia All first authors are attending the conference if you want to discuss with them Link to interactive websites, papers and code below:
2
8
15
LLMs have convincingly demonstrated that coding is the easiest activity, maths is medium hard, and having taste is the hardest
237
2K
31K
My Lab at the University of Edinburgh🇬🇧 has funded PhD positions for this cycle! We study the computational principles of how people learn, reason, and communicate. It's a new lab, and you will be playing a big role in shaping its culture and foundations. Spread the words!
9
82
322
🇨🇦 Heading to @RL_Conference next week to present HyperMARL (CoCoMARL workshop) and Remember Markov (@RLFrameWorkshop) . If you are around, hmu, happy to chat about Multi-Agent Systems (MARL, agentic systems), open-endedness, environments, or anything related! 🎉
📜🤖 Can a shared multi-agent RL policy support both specialised & homogeneous team behaviours -- without changing the learning objective, requiring preset diversity levels or sequential updates? Our preprint “HyperMARL: Adaptive Hypernetworks for Multi-Agent RL” explores this!
1
5
38
happening now!
At #CogSci2025 and curious about resource-rational models of social cognition? Come to Nob Hill A at 11:14 tomorrow to hear me talk about work with @TadegQuillien where we use the information bottleneck to study stereotype use and the outgroup homogeneity bias!
0
0
3
At #CogSci2025 and curious about resource-rational models of social cognition? Come to Nob Hill A at 11:14 tomorrow to hear me talk about work with @TadegQuillien where we use the information bottleneck to study stereotype use and the outgroup homogeneity bias!
1
1
7
very elegant paper!
Our paper on the logic of guesses is now out (w/ @NeilBramley and Chris Lucas). We provide a new information-theoretic perspective on many phenomena (old and new) in judgment under uncertainty. 🧵
0
1
3
Extended Flow-#Lenia paper is out in @ARTL_journal , with more experiments and insights! See 🧵 from 1st-author @eplantec below for a summary. For a full overview of @FlowersINRIA research in Artificial Life, here is a talk I gave at @Demeco2025 : https://t.co/vUXu4EjfCf
Synthetic ecosystems that autonomously and continuously evolve in silico 👾 An Alifer dream we pursue with Flow-Lenia ! If you are interested in complex systems with (1) emergent creatures and (2) intrinsic evolutionary dynamics, go check our new paper ! A 🧵
0
5
20
You work on RL from pixels, and you're tired to wait 10 hours for a DMC run to finish? Or up to 100 hours, if you add video distractors? Well, we got you covered : PixelBrax can run your continuous control experiments from pixels in < 1 hr! Come chat with @trevormcinroe and I at
1
3
16
I wrote a short blog post about our experimental work on the logic of guesses:
1
5
29
Very grateful to @EvostarConf and my peerless coauthors for this! Glad our work resonated and hope to do more along these lines in the future :)
Our paper received the EvoApps best paper AND the EvoStar best student paper @EvostarConf :) Congrats to @MaxTaylorDavie1 @hamongautier @TimotheBoulet from @FlowersINRIA !! Emergent Kin Selection of Altruistic Feeding via Non-Episodic Neuroevolution https://t.co/BUUslHeVrH
0
2
10
Our #CogSci2025 paper led by Madeleine Horner (with Adam Moore) explores people's lay conception of empathy. Main finding: people have a robust expectation that agents who feel empathy are more likely to help, especially when helping is costly.
2
5
11
For anyone at @EvostarConf, this is happening at 14:45 in room A!
@MaxTaylorDavie1 will present our new paper at the @EvostarConf on Thu. Emergent Kin Selection of Altruistic Feeding via Non-Episodic Neuroevolution https://t.co/BUUslHeVrH With @hamongautier and @TimotheBoulet , leveraging previous work with @nisioti_eleni Summary:
1
1
5
If I gave you these evolution rules, can you tell me what will stay constant during the evolution? Very belated thread on recent work on hydrodynamics of circuit systems! Work with Friedrich Hübner, Juan P. Garrahan, and Benjamin Doyon. Preprint: https://t.co/lqJkogiE9B (1/7)
1
1
5
very excited about this line of work, and had a lot of fun collaborating with @TadegQuillien!
Our new paper with @MaxTaylorDavie1 introduces a resource-rational model of Theory of Mind. The model can explain many of the successes and failures of mindreading in human adults and children, and non-human primates: 🧵
1
1
13
Our new paper with @MaxTaylorDavie1 introduces a resource-rational model of Theory of Mind. The model can explain many of the successes and failures of mindreading in human adults and children, and non-human primates: 🧵
2
28
83
We often assume that specialized roles improve performance in multi-agent systems, but when does specialization emerge based on a given task and environment? 🧵👇 ⭐️ New preprint w/ Ruaridh Mon-Williams, @NeilBramley, Chris Lucas, @natvelali, @cocosci_lab
arxiv.org
When should we encourage specialization in multi-agent systems versus train generalists that perform the entire task independently? We propose that specialization largely depends on task...
1
8
37
Me: Your cell phone is a computer Them: Oh yeah, common metaphor. They used to think everything was like pumps and water, then when clockwork was the big technology they thought everything was clockwork, and now the technology is computers they think everything is a computers
80
534
18K
Happy to have this out. Compared to the preprint, we have new quantitative results on the scalings of displacement in the 1D case, where the KPZ term ~t^⅔ and delta-potential term ~t compete. Delta-potential always wins eventually!
Scientists analyzed a random walk from noisy images of a planted directed #polymer, showing that a 1D walker does not exhibit a phase transition. Directed polymers have a transition with decreasing signal to noise where effective inference is impossible. https://t.co/H3rrS0vzXV
0
1
9