Patrick Mineault
@patrickmineault.bsky.social
NeuroAI, vision, open science. NeuroAI researcher at Amaranth Foundation. Previously engineer @ Google, Meta, Mila. Updates from http://neuroai.science
I had way too much fun making these diagrams. Sure, it's not as legible as if I had typed out the text; but I feel like these have a bit more personality
November 6, 2025 at 11:02 PM
I had way too much fun making these diagrams. Sure, it's not as legible as if I had typed out the text; but I feel like these have a bit more personality
Woo! Thanks for subscribing, all!
June 2, 2025 at 5:17 PM
Woo! Thanks for subscribing, all!
Excellent talk from Zico Kolter at ICLR on AI safety. Science of deep learning and AI safety are some of the most impactful ways academic AI researchers can bend the curve
April 24, 2025 at 4:51 PM
Excellent talk from Zico Kolter at ICLR on AI safety. Science of deep learning and AI safety are some of the most impactful ways academic AI researchers can bend the curve
I asked o3 for help with LaTeX quote boxes. This is the example quote it used. The machines are on to us.
April 18, 2025 at 4:45 PM
I asked o3 for help with LaTeX quote boxes. This is the example quote it used. The machines are on to us.
Thanks, we re-use some of this data in the third meta-plot above. Interestingly, scaling curves match qualitatively and relatively well across different groups. See section 1 where we derive a functional form for the scaling laws that work across sims and Lurz et al. arxiv.org/abs/2411.18526
April 18, 2025 at 3:28 PM
Thanks, we re-use some of this data in the third meta-plot above. Interestingly, scaling curves match qualitatively and relatively well across different groups. See section 1 where we derive a functional form for the scaling laws that work across sims and Lurz et al. arxiv.org/abs/2411.18526
But number go up good!
April 17, 2025 at 10:30 PM
But number go up good!
Crowdsourcing this: what are the best examples of scaling laws in foundation models for neuroscience? My top 3 examples: Ctrl+Labs, Sato et al., Mineault et al. @colehurwitz.bsky.social , @mehdiazabou.bsky.social , @nandahkrishna.bsky.social , @averyryoo.bsky.social, @tyrellturing.bsky.social ?
April 17, 2025 at 8:47 PM
Crowdsourcing this: what are the best examples of scaling laws in foundation models for neuroscience? My top 3 examples: Ctrl+Labs, Sato et al., Mineault et al. @colehurwitz.bsky.social , @mehdiazabou.bsky.social , @nandahkrishna.bsky.social , @averyryoo.bsky.social, @tyrellturing.bsky.social ?
The Foundation Model crew at #COSYNE2025. With @josueortc.bsky.social , @colehurwitz.bsky.social , @averyryoo.bsky.social , @nandahkrishna.bsky.social , @auschulz.bsky.social
April 17, 2025 at 3:17 AM
The Foundation Model crew at #COSYNE2025. With @josueortc.bsky.social , @colehurwitz.bsky.social , @averyryoo.bsky.social , @nandahkrishna.bsky.social , @auschulz.bsky.social
Scott Linderman and I ended up on the same panel at #ODIN2025 at the @alleninstitute.bsky.social , wearing the same J Crew shirt. Twins!
April 12, 2025 at 9:20 PM
Scott Linderman and I ended up on the same panel at #ODIN2025 at the @alleninstitute.bsky.social , wearing the same J Crew shirt. Twins!
Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems. This paper looks wild: 264 pages and 1416 references. arxiv.org/abs/2504.01990
April 4, 2025 at 7:41 PM
Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems. This paper looks wild: 264 pages and 1416 references. arxiv.org/abs/2504.01990
I am fully aligned with Eva and Blake's vision. I published this piece a couple of weeks ago: www.neuroai.science/p/what-are-f...
March 28, 2025 at 4:17 PM
I am fully aligned with Eva and Blake's vision. I published this piece a couple of weeks ago: www.neuroai.science/p/what-are-f...
I like to use a diagram on the second slide of my presentations to help people make sense of my unusual background. Unfortunately, I ran out of space with the old one, and 5 circle Venn diagrams are unwieldy. How about this one?
March 27, 2025 at 11:35 PM
I like to use a diagram on the second slide of my presentations to help people make sense of my unusual background. Unfortunately, I ran out of space with the old one, and 5 circle Venn diagrams are unwieldy. How about this one?
Easy hack to have more light in your place: a corn light (100W LED array, equivalent to 1000W incandescent bulb), inside of a photographic light box. It gives about as much light as what I can get from natural sunlight in my house on a sunny day. Will set you back about 60$.
March 12, 2025 at 5:59 PM
Easy hack to have more light in your place: a corn light (100W LED array, equivalent to 1000W incandescent bulb), inside of a photographic light box. It gives about as much light as what I can get from natural sunlight in my house on a sunny day. Will set you back about 60$.
Sometimes it's hard to keep track of all the balls you have up in the air. I made a little script to scrape my daily browser history and email subject lines to get a daily digest. Spaced repetition!
February 17, 2025 at 7:33 PM
Sometimes it's hard to keep track of all the balls you have up in the air. I made a little script to scrape my daily browser history and email subject lines to get a daily digest. Spaced repetition!
Still hasn't cracked pelican on bicycle svg
January 31, 2025 at 8:40 PM
Still hasn't cracked pelican on bicycle svg
o3-mini is here. I guess they got pulled up the release after DeepSeek took over a news cycle
January 31, 2025 at 8:35 PM
o3-mini is here. I guess they got pulled up the release after DeepSeek took over a news cycle
I know I write (overly) long blog posts, but if you scroll down enough you get gems like this
January 27, 2025 at 5:44 PM
I know I write (overly) long blog posts, but if you scroll down enough you get gems like this
I can't wait to see real AI reasoning agents with search, in the line of o1 and DeepSeek-R1. It seems like a step toward building what @michaelnielsen.bsky.social described in his vision of metascience. We might be only a few months away from that scienceplusplus.org/metascience/...
January 23, 2025 at 4:52 PM
I can't wait to see real AI reasoning agents with search, in the line of o1 and DeepSeek-R1. It seems like a step toward building what @michaelnielsen.bsky.social described in his vision of metascience. We might be only a few months away from that scienceplusplus.org/metascience/...
Not sure the exact form it would take, but I would like to create some hybrid of a dashboard, an awesome guide or a map of the field e.g. github.com/analyticalmo...
January 23, 2025 at 4:35 PM
Not sure the exact form it would take, but I would like to create some hybrid of a dashboard, an awesome guide or a map of the field e.g. github.com/analyticalmo...
Halfway through reading How Not to Study a Disease, about how the amyloid hypothesis has dominated Alzheimer's disease research, this interview from @eboyden3.bsky.social hits hard.
January 14, 2025 at 12:39 AM
Halfway through reading How Not to Study a Disease, about how the amyloid hypothesis has dominated Alzheimer's disease research, this interview from @eboyden3.bsky.social hits hard.
The secret to happiness is low expectations 😎 Fun paper on reward prediction errors and subjective well-being www.pnas.org/doi/10.1073/...
January 13, 2025 at 4:49 AM
The secret to happiness is low expectations 😎 Fun paper on reward prediction errors and subjective well-being www.pnas.org/doi/10.1073/...
Vision brain scores have not moved much over the last 6 years. Many models with different loss functions can account for about the same variance (e.g. German-to-English vs predictive masked language modeling for language). 5/
December 12, 2024 at 6:03 PM
Vision brain scores have not moved much over the last 6 years. Many models with different loss functions can account for about the same variance (e.g. German-to-English vs predictive masked language modeling for language). 5/