1/
Our paper is published! Nature Human Behaviour tinyurl.com/p28jy3bx

Driving and suppressing brain activity in the human language network with model (GPT)-selected stimuli

W. Aalok Sathe आलोक साठे Shashank Srikant Maya Taliaferro Mingye (Christina) Wang Martin Schrimpf cvnlab Ev (like in 'evidence', not Eve) Fedorenko 🇺🇦

CCN is hiring summer research interns! Undergrad, grad & pre-doc level positions are open in our research groups, which include computational vision, neural circuits & algorithms, neuroAI and geometric data analysis, + statistical analysis of neural data. apply.interfolio.com/137388

How well do #NeuralNetwork models account for #AuditoryCortex responses? Greta Tuckute Jenelle Feather Dr. Dana Boebinger & Josh McDermott show that most trained DNNs show improvement over a classic baseline model, with cortex-like organization #PLOSBiology plos.io/46Z5ArK

A brief break in #Neurips2023 content to announce that our work on investigating deep neural network models of auditory cortex is out today in PLOS Biology!!!
👂🧠👩‍🔬
Paper: journals.plos.org/plosbiology/ar…
Code: github.com/gretatuckute/a…

Just arrived in New Orleans for #NeurIPS2023 this week.

Topics I am excited about: neuro-AI, neural manifolds (representation geometry), stat physics for machine learning, interpretability, relational & causal representations

My group is presenting their awesome work👇 (1/n)

At #NeurIPS2023 ? Interested in brains, neural networks, and geometry? Come by our **Spotlight Poster** Tuesday @ 5:15PM (#1914) on A Spectral Theory of Neural Prediction and Alignment.
w/ Abdulkadir Canatar SueYeon Chung Albert Wakhloo

Testing artificial neural networks with stimuli whose activations are matched to those of a natural stimulus ('model metamers') demonstrates discrepancies between human and model sensory systems

Jenelle Feather @joshhmcdermott Aleksander Madry et al.

nature.com/articles/s4159…

New paper from our lab, by Jenelle Feather. Asks whether the invariances contained in contemporary artificial neural networks mirror those of humans. See Jenelle’s thread for some of the key ideas and results. Aleksander Madry Guillaume Leclerc MIT Department of Brain and Cognitive Sciences McGovern Institute cbmm MIT Science

Images that humans perceive as completely unrelated can be classified as the same by computational models, according to a new Nature Neuroscience study by Josh McDermott Jenelle Feather Guillaume Leclerc Aleksander Madry MIT Brain and Cognitive Sciences MIT Science

mcgovern.mit.edu/2023/10/16/stu…

This work from NeurIPS 2019 was the start of a much larger project that is now out in Nature Neuroscience! Check it out here: nature.com/articles/s4159…