←
Tom George346 posts
Tom George avatar

Tom George

@TomNotGeorge

Neuroscience/ML PhD @UCL. NeuroAI, navigation, hippocampus

◷ Joined January 2022▧ 67 Photos and videos
Tom George avatar
Tom George@TomNotGeorge·13 Nov 2024

What are the brain’s “real” tuning curves? Our new preprint "SIMPL: Scalable and hassle-free optimisation of neural representations from behaviour” argues that existing techniques for latent variable discovery are lacking. We suggest a much simpl-er way to do things. 1/21🧵

Nanda H Krishna avatar
⟳ Tom George retweeted
Nanda H Krishna@nandahkrishna·Jul 22

What if spike-tokenising decoders could learn from unlabelled neural data? 🧠⚡ Introducing MOJO, a joint SSL+SL framework for POYO- & POSSM-style models. MOJO improves decoding & few-shot transfer and yields more interpretable unit embeddings across neural datasets & tasks. 1/🧵 x.com/nandahkrishna/status/2…

Tom George avatar
Tom George@TomNotGeorge·Apr 3

Great work and a relevant read for neuroscientists and machine learners alike. Made me fundamentally reconsider how I thought about layer normalisation! Well done @RoyEyono

How do neural circuits in the brain implement normalization? 🧠 In our new paper, we show that just normalizing sensory input isn't enough. Crucially, we must also normalize the error signals! 🧵👇 Paper: arxiv.org/abs/2603.17676

Post attachment
Tom George avatar
Tom George@TomNotGeorge·14 Mar 2025

A great article from SWC on our neural data analysis technique (now accepted into ICLR). More updates to come soon...plus will be @CosyneMeeting presenting this too!

SWC@SWC_Neuro13 Mar 2025

How does the brain represent imagined locations? Researchers at SWC, @GatsbyUCL and @UCL developed SIMPL, a method to refine neural tuning curves by correcting distortions from imagined locations—sharpening our view of place cell activity. Read more: sainsburywellcome.org/web/bl…

Post attachment
Tom George avatar
Tom George@TomNotGeorge·6 Jan 2025

🌍🧠💻Applications are well and truly open for the third CaMinA. African nationals studying biology, medicine, engineering, maths etc. can, and should, apply for this summer school in Zambia. Neuro and ML are changing the world and now I the time to get into them, please RT!

🎉Happy New Year! Start 2025 by investing in your future!🚀 Just 2 weeks left to apply for our Computational Neuroscience & Machine Learning course🧠🤖 Let’s make this year one for growth & discovery. RT to spread the word!🙌 🔗trendinafrica.org/trend-cami… #Growth #Opportunity

Tom George avatar
Tom George@TomNotGeorge·25 Nov 2024

CaMinA is back for its 3rd year...this time we're going to beautiful Zambia🇿🇲! I'm proud to see this course grow and bring together the smartest students across Africa with leading neuro institutes like @AllenInstitute and @SWC_Neuro Applications open soon, please share widely!

Post attachment

🌍Exciting news! The 2025 TReND-CaMinA Course will be held in Lusaka, Zambia 🇿🇲 from July 7th–23rd. Dive into computational neuroscience and machine learning with us! 📅 Applications open: December 15th 🔗More info: trendinafrica.org/trend-cami… Stay tuned & spread the word! 🧠✨

Post attachment
TReND-CaMinA avatar
⟳ Tom George retweeted
TReND-CaMinA@trend_camina·22 Nov 2024

🌍Exciting news! The 2025 TReND-CaMinA Course will be held in Lusaka, Zambia 🇿🇲 from July 7th–23rd. Dive into computational neuroscience and machine learning with us! 📅 Applications open: December 15th 🔗More info: trendinafrica.org/trend-cami… Stay tuned & spread the word! 🧠✨

Post attachment
Tom George avatar
Tom George@TomNotGeorge·13 Nov 2024

This isn’t cheating, behaviour has always been there for the taking and we should exploit it. If we ignore behaviour and initialise randomly SIMPL still works but the latent space isn’t smooth and “identifiable”. This is certainly something to consider…. 20/21

Tom George avatar
Tom George@TomNotGeorge·13 Nov 2024

Initialising at behaviour is a powerful trick here. In many regions (e.g., but not limited to, hippocampus 👀), a behavioural correlate (position👀) exists which is VERY CLOSE to the true latent. Starting right next to the global maxima help makes optimisation straightforward.

Post attachment
Tom George avatar
Tom George@TomNotGeorge·13 Nov 2024

These non-local dynamics aren’t a new discovery by any means but this is, in our opinion, the correct and quickest way to find them. 18/21

Tom George avatar
Tom George@TomNotGeorge·13 Nov 2024

And there’s cool stuff in the optimised latent too. It mostly tracks behaviour (hippocampus is still mostly a cognitive map) but does occasional big jumps as though the animal is contemplating another location in the environment. 17/21

Post attachment
Tom George avatar
Tom George@TomNotGeorge·13 Nov 2024

Dubious analogy: Using behaviour alone to study neural representations (status quo for hippocampus) is like wearing mittens and trying to a figure out the shape of a delicate statue in the dark. Everything is blurred. 16/21