iiinova — where your digital self keeps evolving

Reposted by Basile Confavreux

Tim Vogels @tpvogels.bsky.social · 22/09/2026

Playing Pong with spikes, dancing rate networks, composite engrams & #ihngrams4engrams. The labs most recent work-in-progress, now on biorxiv.com. Stay tuned for more Pong doi.org/10.64898/202... Dale doi.org/10.64898/202... Composite engrams doi.org/10.64898/202... Ihngrams doi.org/10.64898/202...

Small sub panel of the "Dale" paper's Fig. 1 ,depicting a humanoid dancing with joy. The paper is about how motor cortex can orchestrate a rich and flexible repertoire of network dynamics for rhythmic and goal-directed movements. Computational studies have begun to illuminate the mechanistic origins of this repertoire, but a comprehensive model that can explain the emergence of both transient and self-sustained dynamics is still missing. Condruz et al. show that three simple ingredients: Dalean connectivity, stability, and nonlinear neural responses, suffice to reverse-engineer networks that produce transient, steady-state, and self-sustained periodic activity. A single dynamical principle underlies this repertoire: the interaction of non-normal amplification, inherent to Dalean networks, with neuronal nonlinearity, so to ignite and sustain multi-stable, controllable dynamics. The approach yields entire families of connectivity matrices that require no hand-tuning or learning of weights. Without fitting them to data, these networks reproduce the population-level signatures of motor cortex, implying that the richness of cortical dynamics need not be sculpted by learning, but may emerge from simple biological ingredients. But really, we just feel like dancing with JOY!!

44115

Reposted by Basile Confavreux

Nastya Krouglova @anastasiakrouglova.bsky.social · 22/03/2026

Our paper “Multifidelity Simulation-based Inference for Computationally Expensive Simulators” has been accepted at ICLR 2026! 🥳 We hope this can be a practical solution for anyone analysing and doing inference on computationally expensive simulators. Paper: openreview.net/pdf?id=bj0dc...

15311

Reposted by Basile Confavreux

Andrew Saxe @saxelab.bsky.social · 16/02/2026

Excited to launch Principia, a nonprofit research organisation at the intersection of deep learning theory and AI safety. Our goal is to develop theory for modern machine learning systems that can help us understand complex network behaviors, including those critical for AI safety and alignment. 1

19428

Reposted by Basile Confavreux

willydorrell.bsky.social @willydorrell.bsky.social · 14/02/2026

Cosyne Viewing Parties Visas, costs, care responsibilities, and environmental concerns all limit Cosyne attendance. Luckily, the talks are livestreamed; but watching alone is the high road to an aneurism. Hence: viewing parties! Gather regionally to watch Cosyne talks! More info: shorturl.at/3DHZX.

shorturl.at

Will's Wild Website

0159

Basile Confavreux @bconfavreux.bsky.social · 01/12/2025

Find us at NeurIPS, Thur 4:30 pm #2115! We know networks have to be both plastic and stable but we're used to thinking about computations, such as memory, as additional requirements. Instead, we find that almost all stable & plastic networks display simple memory abilities.

openreview.net

Memory by accident: a theory of learning as a byproduct of network...

Synaptic plasticity is widely considered to be crucial to the brain’s ability to learn throughout life. Decades of theoretical work have therefore been invested in deriving and designing...

1182

Basile Confavreux @bconfavreux.bsky.social · 04/06/2025

We’ve successfully automated part of the (neuro)scientific process. Now I may be out of a job. After a lot of iterations, here’s our framework for automatic discovery, built to embrace degeneracy and the realities of underconstrained modeling.

2336

Reposted by Basile Confavreux

Tim Vogels @tpvogels.bsky.social · 02/06/2025

We just pushed “Memory by a 1000 rules” onto bioRxiv, where we use clever #ML to find #plasticity quadruplets (EE, EI, IE, II) that learn basic stability in spiking nets. Why is it cool? We find 1000s!! of solutions, and they don’t just stabilise. They #memorise! www.biorxiv.org/content/10.1...

biorxiv.org

Memory by a thousand rules: Automated discovery of functional multi-type plasticity rules reveals variety & degeneracy at the heart of learning

Synaptic plasticity is the basis of learning and memory, but the link between synaptic changes and neural function remains elusive. Here, we used automated search algorithms to obtain thousands of str...

413747