🔥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

Great honor to work with Akram Bakkour and Daphna on this paper. Be on the lookout for some of our favorite dishes in Figure 1.

Check out our new representation learning algorithm for analyzing animal behavioral videos; we combine the output of supervised pose estimation (e.g. DeepLabCut 🦄) with unsupervised methods (VAE) to extract interpretable behavioral features: biorxiv.org/content/10.110… Thread: 1/7

Check out our new paper about what happens to the options we decide against. Daphna Shohamy Nature Communications If you want the short version, take a look at our blog post: socialsciences.nature.com/posts/how-we-l…

Deciding to choose a certain option over another also impacts our valuation of the *unchosen* option Natalie Biderman Columbia University's Zuckerman Institute Department of Psychology at Columbia University nature.com/articles/s4146… Behind the Paper: go.nature.com/3yakPxA

Cool work from Dan Biderman and team on randomized early truncation.

our PS-VAE paper is out in PLOS Comp Bio. We included a new model variant whose representations are robust to image differences across behavioral sessions, and a discussion of PS-VAE within related ML approaches. Columbia NeuroTheory Columbia University's Zuckerman Institute Link: journals.plos.org/ploscompbiol/a…

Ever noticed when your chosen option turns sour, the unchosen suddenly seems appealing? Daphna Shohamy, Sam Gershman and I have shown that this counterfactual inference is a unique signature of an internal policy-gradient model of choice and it is driven by memory.

amazing work by an amazing team. and i'm not biased at all :)

Check out this cool work by the super team Dan Biderman Avanika Narayan Sabri Eyuboglu

Very proud of this man ❤️