The elegance of ML is the elegance of biology, not the elegance of math or physics. Simple gradient descent creates mind-boggling structure and behavior, just as evolution creates the awe inspiring complexity of nature.

Sometimes it takes a step back—or up—to see the full picture. Take a look at UChicago from a new perspective.

Also.. it costs ~$10k to publish into a nature publishing journal (price is agnostic to daughters, e.g. physics, methods, biotech, etc). This will save so much taxer payer money and give back to graduate researchers, postdocs, PIs, and allow for conducting more

Scientific discovery is already difficult enough... adding escalating political conflict that's leading to modern warfare is heartbreaking. 💔 Especially painful for me to see as someone born during the Bosnian wars and who immigrated to America as a war refugee. Prayers for the

Well said!

What Are Virtual Cells? centuryofbio.com/p/virtual-cell Two months ago, I started working on an essay to answer this question. The goal was to cover a few of the recent research results. Simple as that! Instead, I went down a rabbit hole exploring ideas around cellular simulation

The West Side of Chicago's neighborhood fireworks as seen from my 5th story rooftop with a 600mm zoom.

No City fireworks this year? Chicago residents say, "Hold my beer". #chicago

Nope..

Exactly why science is so addictive & fulfilling regardless of all the BS overhead issues that it takes to conduct science ♥️

🙌🙌🙌

I drew this whirlpool in an ocean with mathematical equations.

We wouldn't have to worry about high fructose corn syrup if synthetic biology could deliver sweet de novo proteins that are just as sweet as traditional sugar, providing the same satisfying dopamine release, but in a healthier, more controlled way.

'water is transparent only within a very narrow band of the electromagnetic spectrum, so living organisms evolved sensitivity to that band, and that's what we now call "visible light". ' (found via HN)

Happy summer 🌞 ❤️ The University of Chicago Chicago Booth

Lecleeeerrrrrc 🏎️

Totally accurate post/observation! However, why should we expect scaling pretraining would allow for increased mutational effect prediction? A similar trend, LLM in natural text shows that scaling pretraining models don't necessarily allow them to behave like human intentions and

This is so cool! Starting to believe information bits is the most fundamental aspect of the universe over particles/atoms/electrons.

There is a reason why it's been 98 years. E.coli (+others) filled river 🤢💀

This reminds me of John Wick..