Lightnews — Scholar-powered news

Andrew Saxe

@saxelab.bsky.social

4.1K followers 470 following 30 posts

Professor at the Gatsby Unit and Sainsbury Wellcome Centre, UCL, trying to figure out how we learn

Posts Media Videos Starter Packs

Pinned

Andrew Saxe @saxelab.bsky.social · Jul 14

Excited to share new work @icmlconf.bsky.social by Loek van Rossem exploring the development of computational algorithms in recurrent neural networks.

Hear it live tomorrow, Oral 1D, Tues 15 Jul West Exhibition Hall C: icml.cc/virtual/2025...

Paper: openreview.net/forum?id=3go...

(1/11)

ICML Poster Algorithm Development in Neural Networks: Insights from the Streaming Parity TaskICML 2025

icml.cc

1 11 39

Reposted by Andrew Saxe

Charlotte Volk @charlottevolk.bsky.social · 9d

🚨 New preprint alert!

🧠🤖
We propose a theory of how learning curriculum affects generalization through neural population dimensionality. Learning curriculum is a determining factor of neural dimensionality - where you start from determines where you end up.
🧠📈

A 🧵:

tinyurl.com/yr8tawj3

The curriculum effect in visual learning: the role of readout dimensionality

Generalization of visual perceptual learning (VPL) to unseen conditions varies across tasks. Previous work suggests that training curriculum may be integral to generalization, yet a theoretical explan...

tinyurl.com

1 22 69

Reposted by Andrew Saxe

Gatsby Computational Neuroscience Unit @gatsbyucl.bsky.social · 14d

🙋Are you interested in bridging theory & experiments?

Applications are now open for 2026 entry to the Gatsby Unit & SWC joint PhD programme.

Join us and be part of a vibrant research community!

💰 Fully-funded 4-year programme
ℹ️ www.ucl.ac.uk/life-science...

@sainsburywellcome.bsky.social

10 14

Reposted by Andrew Saxe

Laureline Logiaco @laurelinelogiaco.bsky.social · 12d

Interested in doing a Ph.D. to work on building models of the brain/behavior? Consider applying to graduate schools at CU Anschutz:
1. Neuroscience www.cuanschutz.edu/graduate-pro...
2. Bioengineering engineering.ucdenver.edu/bioengineeri...

You could work with several comp neuro PIs, including me.

1 30 51

Reposted by Andrew Saxe

Kris Jensen @kristorpjensen.bsky.social · 15d

I’m super excited to finally put my recent work with @behrenstimb.bsky.social on bioRxiv, where we develop a new mechanistic theory of how PFC structures adaptive behaviour using attractor dynamics in space and time!

www.biorxiv.org/content/10.1...

9 85 200

Reposted by Andrew Saxe

Gatsby Computational Neuroscience Unit @gatsbyucl.bsky.social · 16d

Interested in a #PhD in machine learning & theoretical neuroscience?

Applications for 2026 entry to the Gatsby Unit PhD programme are open!

💰 Fully-funded 4-year programme
🤝 Close links to @sainsburywellcome.bsky.social
ℹ️ www.ucl.ac.uk/life-sciences/gatsby/study-and-work/gatsby-unit-phd-programme

1 6 23

Reposted by Andrew Saxe

Dirk Gütlin @gutlin.bsky.social · 17d

Structure transfer and consolidation in visual implicit learning

🧠🟦 🧠🤖

elifesciences.org/articles/100...

Structure transfer and consolidation in visual implicit learning

Sleep is essential for consolidating implicitly acquired perceptual knowledge that enables the knowledge transfer effect via newly learned structured information observed in prior studies of explicit ...

elifesciences.org

5 13

Reposted by Andrew Saxe

Abhilasha Joshi, PhD @rhythmicspikes.bsky.social · 22d

1/
🚨 New preprint! 🚨

Excited and proud (& a little nervous 😅) to share our latest work on the importance of #theta-timescale spiking during #locomotion in #learning. If you care about how organisms learn, buckle up. 🧵👇

📄 www.biorxiv.org/content/10.1...
💻 code + data 🔗 below 🤩

#neuroskyence

9 56 120

Reposted by Andrew Saxe

Chris Baldassano @chrisbaldassano.bsky.social · 23d

What happens when we learn a new shortcut between places we thought were unconnected? Hannah found that the hippocampus rapidly adjusts its representations of environments to join them into a connected map - excited to share this final paper from her PhD work with me and @mariamaly.bsky.social !

Mariam Aly @mariamaly.bsky.social · 23d

How do we update our predictions when our environment changes?

The hippocampus rapidly integrates previously distinct sequences to support updated predictions.

Proud of this work with Hannah Tarder-Stoll & @chrisbaldassano.bsky.social!

www.biorxiv.org/content/10.1...

The Hippocampus Rapidly Integrates Sequence Representations During Novel Multistep Predictions

Memories for temporally extended sequences can be used adaptively to predict future events on multiple timescales, a function that relies on the hippocampus. For such predictions to be useful, they sh...

www.biorxiv.org

2 10 40

Reposted by Andrew Saxe

Dan Yamins @dyamins.bsky.social · 23d

Here is our best thinking about how to make world models. I would apologize for it being a massive 40-page behemoth, but it's worth reading. arxiv.org/pdf/2509.09737

arxiv.org

2 17 70

Reposted by Andrew Saxe

Joey Rudoler @jrudoler.bsky.social · Sep 7

Just spent two wonderful weeks in London for the Analytical Connectionism Summer School (hosted at Gatsby/UCL this year). Met lots of wonderful scientists at the intersection of cog neuro and machine learning. Learned a lot and can’t recommend more highly! Small meetings rule

1 1 9

Reposted by Andrew Saxe

Joao Barbosa @jbarbosa.org · Sep 6

I recently learned: w/ lesioned 8A, you can do many WM tasks but not one👇

Guess what happens when you decoding from 8A during each of these tasks? They are all the same.

Decoding is like a quality check, it provides almost no info about function

scholar.google.com/citations?vi...

2 1 6

Reposted by Andrew Saxe

Timothy O'Leary @timothyoleary.bsky.social · Sep 6

We all know that correlation doesn't imply causation. So we took some correlations and tested if they were causal. Here's what happened:

www.cell.com/cell-reports...

An optical brain-machine interface reveals a causal role of posterior parietal cortex in goal-directed navigation

Relating neural circuitry to behavior is challenging due to closed loop interactions between neural activity, actions, and sensations. Sorrell et al. present evidence for a causal role of mouse PPC in...

www.cell.com

3 11 45

Reposted by Andrew Saxe

Erin Grant @eringrant.me · Aug 13

Our #CCN2025 GAC debate w/ @gretatuckute.bsky.social, Gemma Roig (www.cvai.cs.uni-frankfurt.de), Jacqueline Gottlieb (gottlieblab.com), Klaus Oberauer, @mschrimpf.bsky.social &‬ @brittawestner.bsky.social asks:

📊 What benchmarks are useful for cognitive science? 💭
2025.ccneuro.org/gac

Speakers and organizers of the GAC debate. Time and location of the GAC debate: 5 PM in Room C1.03.

1 16 50

Reposted by Andrew Saxe

Erin Grant @eringrant.me · Aug 13

Are similar representations in neural nets evidence of shared computation? In new theory work w/ Lukas Braun (lukasbraun.com) & @saxelab.bsky.social, we prove that representational comparisons are ill-posed in general, unless networks are efficient.

@icmlconf.bsky.social @cogcompneuro.bsky.social

3 20 72

Reposted by Andrew Saxe

Denis Lan @denislan.bsky.social · Aug 7

My first PhD paper - with @lhuntneuro.bsky.social and @summerfieldlab.bsky.social - is now out in @plosbiology.org! We ask: how do humans (and deep neural networks) navigate flexibly even in unfamiliar environments, such as a new city? Link: plos.io/45uSwNm 🧵 (1/6)

Cartoon image of me looking at a map, with a stadium behind me and a hotel and ferris wheel across the river in the background. I am thinking about going to the ferris wheel

1 11 30

Reposted by Andrew Saxe

Guido Meijer @guidomeijer.com · Aug 5

🚨Pre-print alert🚨

We stimulated serotonin with optogenetics while doing large-scale Neuropixel recordings across the mouse brain. We found strong widespread modulation of neural activity, but no effect on the choices of the mouse 🐭

How is this possible? Strap in! (1/9) 👇🧵

doi.org/10.1101/2025...

Serotonin drives choice-independent reconfiguration of distributed neural activity

Serotonin (5-HT) is a central neuromodulator which is implicated in, amongst other functions, cognitive flexibility. 5-HT is released from the dorsal raphe nucleus (DRN) throughout nearly the entire f...

doi.org

3 35 91

Reposted by Andrew Saxe

Juan Gallego @juangallego.bsky.social · Aug 6

Very happy about my former mentor Sara Solla having received the Valentin Braitenberg Award for her lifelong contributions to computational neuroscience!

Sara will be giving a lecture at the upcoming @bernsteinneuro.bsky.social meeting which you shouldn't miss.

bernstein-network.de/en/newsroom/...

Sara A. Solla receives the Valentin Braitenberg Award for Computational Neuroscience 2025 – Bernstein Network Computational Neuroscience

bernstein-network.de

1 13 73

Reposted by Andrew Saxe

Sainsbury Wellcome Centre @sainsburywellcome.bsky.social · Aug 6

12 leading neuroscientists tackle a big question: Will we ever understand the brain?

Their reflections span philosophy, complexity, and the limits of scientific explanation.

www.sainsburywellcome.org/web/blog/wil...

Illustration by @gilcosta.bsky.social & @joanagcc.bsky.social

Illustration of the brain in blue and yellow

2 3 15

Reposted by Andrew Saxe

David Sussillo @sussillodavid.bsky.social · Aug 4

Coming March 17, 2026!
Just got my advance copy of Emergence — a memoir about growing up in group homes and somehow ending up in neuroscience and AI. It’s personal, it’s scientific, and it’s been a wild thing to write. Grateful and excited to share it soon.

8 37 180

Reposted by Andrew Saxe

Laurence Hunt @lhuntneuro.bsky.social · Aug 2

Our new paper is out! When navigating through an environment, how do we combine our general sense of direction with known landmark states? To explore this, @denislan.bsky.social used a task that allowed subjects (or neural networks) to choose either their next action or next state at each step.

PLOS Biology @plosbiology.org · Aug 1

How do humans navigate unfamiliar environments? @denislan.bsky.social @lhuntneuro.bsky.social @summerfieldlab.bsky.social show that humans & deep meta-learning networks combine ‘vector-based’ & ‘transition-based’ strategies for flexible navigation in similar ways @plosbiology.org 🧪 plos.io/45uSwNm

Task design and experimental set-up. Top left: underlying structure of the 8 × 8 grid, unseen by participants. Every state is represented by an image of an object, and these objects and their positions change on every trial. Top right: schematic diagram of the ‘map reading’ phase of each trial. Participants see a top–down view of the grid with objects obscured and successively click on blue squares to reveal ‘landmark’ objects at the location. After 16 clicks have been completed, a yellow square appears. Clicking on the yellow square reveals the ‘goal’ object for the trial. Bottom: schematic diagram of the navigation phase of each trial. Participants start in a random, previously unobserved location and are tasked with navigating to the ‘goal’ object they had just learnt about (displayed at the top). They can navigate in two ways. First, they could choose a direction to travel in by clicking on the corresponding arrow (highlighted yellow). This is analogous to using a ‘vector-based’ strategy. Alternatively, they could choose an adjacent state to travel to by clicking on one of the associated images (displayed in a random order; highlighted blue). This corresponds to using a ‘transition-based’ navigation strategy.

25 70

Reposted by Andrew Saxe

Clementine Domine 🍊 @CCN @clementinedomine.bsky.social · Jul 29

🎓Thrilled to share I’ve officially defended my PhD!🥳

At @gatsbyucl.bsky.social, my research explored how prior knowledge shapes neural representations.

I’m deeply grateful to my mentors, @saxelab.bsky.social and @caswell.bsky.social, my incredible collaborators, and everyone who supported me!

5 2 38

Reposted by Andrew Saxe

Claudia Clopath Lab @clopathlab.bsky.social · Jul 23

Trying to train RNNs in a biol plausible (local) way? Well, try our new method using predictive alignment. Paper just out in Nat. Com. Toshitake Asabuki deserves all the credit!
www.nature.com/articles/s41...

www.nature.com

1 16 56

Reposted by Andrew Saxe

Gatsby Computational Neuroscience Unit @gatsbyucl.bsky.social · Jul 21

🥳 Congratulations to Rodrigo Carrasco-Davison on passing his PhD viva with minor corrections! 🎉

📜 Principles of Optimal Learning Control in Biological and Artificial Agents.

1 16

Reposted by Andrew Saxe

Matteo Carandini @carandinilab.net · Jul 18

Welcome to Bluesky @kenneth-harris.bsky.social !

3 20

Reposted by Andrew Saxe

Athena Akrami @athenaakrami.bsky.social · Jul 18

🎉 Heron is finally out @elife.bsky.social! Led by George Dimitriadis, with Ella Svahn & @macaskillaf.bsky.social

🧪 🧠 🐭 🤖

If you wonder why yet another tool for experimental pipelines, read the 🧵 below:

#neuroscience #neuroskyence #OpenSource

1/
elifesciences.org/articles/91915

1 12 38