🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

From logical reasoning to mental simulation, recurrent neural networks (RNNs) possess an incredible capacity for computation. To fully harness this capacity, we developed a programming language for RNNs arxiv.org/abs/2203.05032. Here is a game of pong programmed inside of an RNN.

Interested in doing a postdoc in Heidelberg/Mannheim Health + Life Science Alliance Heidelberg Mannheim investigating cognitive maps of social interactions and mental health? Check out and apply for the project proposed by Christoph Korn and me: health-life-sciences.de/jobs/interdisc… #OpenScience #neuroscience #mentalhealth

In the workshop on the creative scientific process at NYU, we discussed how to search for new questions (particularly by finding contradictions), how to use anthropomorphisms and interdisciplinary thinking to generate new ideas, and how hypotheses can be a liability.🧵of 🧵's ⬇️

Epilepsy is more than seizures, and here we map the organization of language and working memory systems in frontal and temporal lobe epilepsy (FLE, TLE) –now in press Brain! academic.oup.com/brain/advance-… mentored by Matthias Koepp Dani S. Bassett @BorisBernhardt 1/1

New computational tools to build bridges between small data and big data ! nature.com/articles/s4159… Led by Tong He and Thomas Yeo, with Lijun AN | 安丽军 Simon Eickhoff @AvramHolmes and others.

Null models in network neuroscience Nature Rev Neurosci w/ Bratislav Misic | nature.com/articles/s4158… | Null models help disentangle dependencies between network features, revealing to what extent the presence or magnitude of a feature of interest arises as a consequence of other features.

Check out our review looking at multimodal network biomarkers of cognitive and affective difficulties in epilepsy! out Network Neuroscience led by Raúl Rodríguez Cruces, with @J_Royer_ , Sara Lariviere Dani S. Bassett @BorisBernhardt – fantastic team!!! 🤩🥳

Highly recommended!

Visit me at my #OHBM2022 poster (MT033) on Monday (June 20th), 12:15-1:15pm, and Tuesday (June 21st), 1-2pm. I'm happy to chat about #TMS personalization, brain connectivity, E-field modelling, and concurrent #TMSfMRI 🧠🧲⚡🎯

Great idea by Georgia Koppe and Janine Thome harvesting latent behavioral models for guiding optimal experimental design!

As one of the first institutions, the CIMH joins the German Reproducibility Network (GRN) and thus strengthens the Open Science approach, which stands for open and reproducible research. #OpenScience 👉zi-mannheim.de/en/institute/n…

A project at the CIMH is investigating the efficacy of oxytocin in combination with various psychosocial forms of treatment for schizophrenia. Urs Braun #mentalhealth 👉zi-mannheim.de/en/institute/n…

Please RT: jobs.ruhr-uni-bochum.de/jobposting/a02… Der Lehrstuhl für eHealth und Sports Analytics der Fakultät für Sportwissenschaft sucht ab dem nächstmöglichen Zeitpunkt eine/n Doktorand*in (m/w/d) für die Dauer von 3 Jahren mit 19,915 Wochenstunden (TVL E13).

Workshop in Osnabrück on October 5-6 & via livestream: We gonna discuss how neurons and artificial neural networks transmit information; what advantages and disadvantages biological and artificial networks have; to what extent needs AI to be biologically plausible; etc...

Interested in how brain structure constrains function? Check our latest PP - a tour de force from James Pang, Kevin Aquino 🧠🔑 & team, incl. Michael Breakspear Ben Fulcher Marianne Oldehinkel & P Robinson. V excited about this work; peep the thread...

Interested in reconstructing computational dynamics from neural data using RNNs? Here we review dynamical systems (DS) concepts, recent #ML/ #AI methods for recovering DS from data, evaluation, interpretation, analysis, & applic. in #Neuroscience: biorxiv.org/content/10.110… A 🧵...

Weight pruning by size is a standard #ML #AI technique to produce sparse models, but in our ICML Conference paper arxiv.org/abs/2406.04934 we find it doesn’t work for learning #DynamicalSystems! Instead, via geometry-based pruning we find *network topology* is far more important! (1/5)

Why does the neural network display modular structure during learning and how does the modularity characterize neurodevelopment and cognitive behavior? Please check our recent paper in Science Advances with Marcelo Mattar. science.org/doi/10.1126/sc…