Iranians are experiencing a collective trauma. Thousands have been killed/injured in recent events, the economy is crippled & the threat of a wider conflict is real. This is especially difficult for those living in Iran, as many have lost (or fear losing) loved ones. www.nature.com/articles/d41...
03.02.2026 14:11
๐ 70
๐ 32
๐ฌ 1
๐ 7
Hats off to the great Anne "I don't do fancy" Treisman!
01.02.2026 21:55
๐ 89
๐ 14
๐ฌ 1
๐ 2
Links:
๐ง My neuro-RHEL abstract:
openreview.net/attachment?i...
๐ Original RHEL paper: openreview.net/forum?id=w1i...
๐ฎ Interactive demo by Guillaume Pourcel to understand RHEL: guillaumepourcel.github.io/rhel-interac...
03.12.2025 00:08
๐ 0
๐ 0
๐ฌ 0
๐ 0
๐ And if you want to hear more (or rather all) about RHEL Guillaume Pourcel will present the full framework in San Diego ๐บ๐ธ on Friday 5:
- Oral at 4:10 pm PST
- Poster from 4:30 pm PST
Maxence Ernoult presents in Copenhagen ๐ฉ๐ฐ at #EurIPS2025:
- Oral on Friday 5 at 9 am
- Poster on Thursday 4
02.12.2025 23:39
๐ 0
๐ 0
๐ฌ 1
๐ 0
๐ค If you're interested in our latest results, comparison with existing approaches or just curious about new theories for learning in the brain, stop by!
02.12.2025 23:39
๐ 0
๐ 0
๐ฌ 1
๐ 0
๐ The key result: Our learning rule is a form of contrastive Hebbian learning and recovers the exact gradients through time!
02.12.2025 23:39
๐ 0
๐ 0
๐ฌ 1
๐ 0
Amazingly there is a new alternative to BPTT in town: RHEL==Recurrent Hamiltonian Echo Learning (Pourcel & Ernoult, 2025)! We took inspiration from Hopfield energy models to design a novel Hamiltonian NN. In our model, neurons are now second-order oscillating ODEs that can be trained using RHEL...
02.12.2025 23:38
๐ 0
๐ 0
๐ฌ 1
๐ 0
๐ข Presenting our work at #NeurIPS2025 ๐บ๐ธ today (6-9 pm) @wimlworkshop.bsky.social!
๐ง The question: How can the brain learn temporal dependencies without an artificial backward pass? BPTT is notoriously harder than BP and is still uncharted territory (Lillicrap & @adamsantoro.bsky.social , 2019).
02.12.2025 23:37
๐ 3
๐ 0
๐ฌ 1
๐ 0
๐ช๐ต๐ฒ๐ฟ๐ฒ ๐ฑ๐ผ ๐ฟ๐ฒ๐๐ฎ๐ฟ๐ฑ๐ (๐๐ฎ๐น๐๐ฒ) ๐ฐ๐ผ๐บ๐ฒ ๐ณ๐ฟ๐ผ๐บ?
Implications for reinforcement learning.
Very interesting paper.
doi.org/10.1016/j.ti...
#neuroskyence
03.09.2025 16:27
๐ 111
๐ 24
๐ฌ 7
๐ 0
What a Swedish Series (Real Humans) Teaches Us About AI Safety
A cool series that brings AI safety concepts to life...
If you are curious about AI safety/alignment or just look for a good series picturing a future with AI at scale, check my post: open.substack.com/pub/aisig/p/...
03.09.2025 08:01
๐ 4
๐ 0
๐ฌ 0
๐ 0
Et je viens de voir passer de strip de @unpied.bsky.social qui rรฉsume pas mal ce qu'on peut penser de cette histoire.
19.06.2025 07:17
๐ 230
๐ 73
๐ฌ 5
๐ 1
The Replicards โ Replicards
With @eliamascolo.bsky.social we developed an interactive class on biological evolution (focusing on mutations ๐พ) for science #teacher through a card game ๐ฒ We crash-tested it with #HighSchool students from Milan and made it available online for broader use: replicards.netlify.app #SciCom #EvoBio
25.02.2025 11:36
๐ 18
๐ 6
๐ฌ 3
๐ 2