Alex Atanasov
@ABAtanasov
Followers
1K
Following
23K
Media
55
Statuses
455
Fascinated by scaling and universality. PhD from @harvardphysics. Not a Bayesian. Opinions my own 🇧🇬
New York
Joined January 2017
If you’re an undergrad interested in shaping the future of AI, you can either fine-tune underwhelming language models at a leading AI lab OR you can join the group of one of the most talented scientists I know, pursuing some of the most promising research at the intersection of.
Excited to announce that I will be joining @UTAustin with a joint position between @OdenInstitute for Computational Science and dept of Neuroscience in FL 2026! I plan on recruiting PhD students and postdocs interested in mathematics of neural computation (more details to come).
0
6
46
RT @dwarkesh_sp: Honestly the thing that motivated me to do this episode was learning that there's less than $200M/year of smart philanthro….
0
208
0
RT @SussilloDavid: Coming March 17, 2026!.Just got my advance copy of Emergence — a memoir about growing up in group homes and somehow endi….
0
37
0
Generate videos in just a few seconds. Try Grok Imagine, free for a limited time.
46
25
272
In addition to being one of the most promising upcoming physicists, he wrote a great paper on a topic I gave him, earning top place at RSI last summer. He explored the dichotomy between rich and lazy training. It’s super accessible and well-written:.
0
0
3
So proud of my RSI mentee Brian (second from left) on securing 1st place for the USA.
The US just won the International Physics Olympiad!. They beat out China to claim the top spot.
1
0
20
This is absolutely the best possible education a self-motivated kid can get.
she was self-taught via khan academy, homeschooled, and had finished calculus at age 11. wow the internet creates the opportunity for such miracles. Thanks to Sal Khan, and the continued work of others who are democratizing knowledge.
0
0
3
RT @ThomasVanRiet2: Another nice IKKT related paper this year. If spacetime can truly and fully emerge from random….
arxiv.org
The IKKT matrix model has been investigated as a promising nonperturbative formulation of superstring theory. One of the recent developments concerning this model is the discovery of the dual...
0
11
0
RT @provisionalidea: as a linguist, the speaker should know better: it’s not about “speaking more efficiently” (using mathematical terms fo….
0
737
0
Losing my mind rn.
0
0
9
RT @KempnerInst: Check out the #KempnerInstitute's presentations at #ICML2025 today!. #AI #NeuroAI #LLMs . 🧵Here's a thread of abstracts.….
0
4
0
Sensational work by the gang.
Great to see this one finally out in PNAS! Asymptotic theory of in-context learning by linear attention Many thanks to my amazing co-authors Yue Lu, @maryiletey, Jacob Zavatone-Veth and @AninditaMaiti7.
0
0
4
Preach.
We don’t have AI self-improves yet, and when we do it will be a game-changer. With more wisdom now compared to the GPT-4 days, it's obvious that it will not be a “fast takeoff”, but rather extremely gradual across many years, probably a decade. The first thing to know is that.
0
0
2
RT @vikhyatk: You can differentiate 90% of the ML yappers on Twitter by asking them to prove the Marchenko–Pastur Law or recite Guerra’s an….
0
7
0
RT @ABAtanasov: @nabla_theta @jxmnop This is demonstrably false. Neural nets in the kernel regime achieve generalization, overfit benignly,….
0
1
0
RT @EugeneVinitsky: There is no AI research program in the US without Chinese and Indian students. If you think otherwise, it’s because you….
0
248
0
Katie’s really doing a service to the community here.
There were so many great replies to this thread, let's do a Part 2!. For scaling laws between loss and compute, where loss = a * flops ^ b + c, which factors change primarily the constant (a) and which factors can actually change the exponent (b)?.
0
0
8
RT @pfau: How is mathematics at Harvard, now that it is free from the antisemitic influence?.
0
21
0
RT @Harvard: Without its international students, Harvard is not Harvard.
harvard.edu
We have just filed a complaint and a motion for a temporary restraining order will follow. As we pursue legal remedies, we will do everything in our power to support our students and scholars.
0
16K
0
RT @_katieeverett: 1. We often observe power laws between loss and compute: loss = a * flops ^ b + c.2. Models are rapidly becoming more ef….
0
92
0
RT @demishassabis: I think about this type of thing _all_ the time, how did we do all this with our hunter gatherer brains. .
0
143
0