Your Brain on Code: What Brains and AI Models Both Learn
Shashank Srikant
NeurIPS · 2022Paper discussed
Using brain recordings to read what AI models of code have learned, and what people represent when they read programs.
Interpretable is a science podcast from equ. Researchers and clinicians walk through their work on AI for health, from pandemics and new medicines to the mind, and every episode links to the paper behind it. Hosted by Dr. Omar Costilla Reyes.
Know someone we should put on the mic? Pitch a guest
Latest:Your Brain on Code: What Brains and AI Models Both Learn38 min
Research featured from
NeurIPSNatureICLRScience
About 5 hours of listening · free, in the browser
00 Why Interpretable
Every episode looks at AI for health through one lens: can people understand what the model is doing? Some episodes are about models built to be readable, some about methods for reading the ones we have, and some about the human side, how people understand data, decisions, and their own minds.
Start with Episode 00001 Fresh episodes
Shashank Srikant
NeurIPS · 2022Paper discussed
Using brain recordings to read what AI models of code have learned, and what people represent when they read programs.
Mark K. Ho
Nature · 2022Paper discussed
A precise, testable model of how people simplify problems before they plan.
Minghao Guo
ICLR · 2022Paper discussed
A generative model for chemistry built from learned rules: valid by design and readable by chemists.
Dr. Jacob E. Lemieux
Science · 2022Paper discussed
An interpretable Bayesian model that turns millions of viral genomes into a fitness estimate for each mutation.
Dr. Omar Costilla Reyes
NeurIPS · 2022Paper discussed
Why scientific knowledge looks like code, and why that makes AI readable.
Dr. Thomas R. Insel
Penguin Press · 2022
Why mental health care needs signals people can understand, and where data and digital biomarkers fit.
02 Pick a thread
03 Made for listening
The player follows you around the site and remembers where you stopped.
Every episode has chapters, so you can skip straight to the idea you came for.
Show notes and links to the papers behind every talk, for when one idea is not enough.
Send a link that starts at the exact minute that made you think.

04 Your host
I'm a researcher in neurosymbolic AI and computational neuroscience, and I lead equ healthcare, where we build privacy-first, explainable AI for health. On the show, I bring you the people doing the work, from tracking a pandemic to designing medicines to understanding the mind, with the papers to back it up.
More about Omar(opens in a new tab)Never miss an episode
One short email when a new episode is out, with the speaker, the papers mentioned, and a link to listen. Prefer a podcast app? Add the RSS feed to any app that supports it.
05 Contact
Working on AI for mental health, public health, drug discovery, or the brain? We would like to hear about it. Press and research collaborations are welcome too.