Often found scribbling down math with intermittent bursts of bashing out code.
New York City, USA
Joined June 2011
- Tweets587
- Following357
- Followers1.7K
- Likes4.4K
Robert M. Gower retweeted
📣 We are excited to announce the workshop
Emerging Directions in Probabilistic Modeling: Methods and Scientific Applications
📅 Oct. 5–7, 2026
📍 Flatiron Institute & IBM Research, NYC
Applications are welcome by Aug. 22:
edpmworkshop.github.io
Robert M. Gower retweeted
I am looking for the amazing postdocs to advance the foundations of optimization, machine learning, and AI.
Requirement:
- Outstanding training and track record in mathematics and computer science. - Stellar research track record in optimization, ML and AI, a closely related field.
Supplementary info: richtarik.org/i_apply.html
Lead your independent research. Create global impact.
Designed for exceptional early-career researchers, KAUST Global Fellowship offers the opportunity to pursue independent research in areas aligned with global priorities, including Health, Water, Energy, Food Security, AI, and Marine Science.
Register your interest: kgfp.kaust.edu.sa/
Robert M. Gower retweeted
Poster today at 5pm! #ICML2026
"It's easier to tune the LR for method A than for B."
We tried to formalize this for model-based stochastic optimization methods.
We find a key quantity, called stability index, that describes how stable a (weakly) convex bound is as a function of LR.
📚arxiv.org/abs/2602.09842
Robert M. Gower retweeted
"It's easier to tune the LR for method A than for B."
We tried to formalize this for model-based stochastic optimization methods.
We find a key quantity, called stability index, that describes how stable a (weakly) convex bound is as a function of LR.
📚arxiv.org/abs/2602.09842
Robert M. Gower retweeted
Check out my poster today (Thurs) at 11am--2pm session. Exhibit Hall C,D,E Poster Location: #602
"Fisher meets Feynman: score-based variational inference with a product of experts" (NeurIPS spotlight)
with @gowerrobert David Blei and Lawrence Saul
@FlatironInst #NeurIPS2025
ALT Fisher meets Feynman: score-based variational inference with a product of experts
Robert M. Gower retweeted
We’re recruiting for both postdoc and open-rank positions. Learn more about ML@CCM 👉 users.flatironinstitute.org/…
I’ll also be in San Diego for NeurIPS — feel free to DM if you’re interested in #AIforScience or #GenerativeAI
Want to do fundamental ML research in NYC? 🧠
The Center for Computational Mathematics @FlatironInst @SimonsFdn is hiring!
– Flatiron Research Fellow (postdoc, by Dec 1): apply.interfolio.com/173401
– Open Rank (by Jan 15): apply.interfolio.com/173640
Robert M. Gower retweeted
Fisher meets Feynman! 🤝
We use score matching and a trick from quantum field theory to make a product-of-experts family both expressive and efficient for variational inference.
To appear as a spotlight @ NeurIPS 2025.
#NeurIPS2025 (link below)
ALT Fisher meets Feynman: score-based variational inference with a product of experts
Robert M. Gower retweeted
Call for participation:
KAUST Workshop on Distributed Training in the Era of Large Models
kaust.edu.sa/events/dtelm25/
location: KAUST, Saudi Arabia
dates: Nov 24-26, 2025.
There will be a chance for some participants to present a poster and/or give a lightning talk.
Robert M. Gower retweeted
Come to HilD tomorrow @ICML2025 !
We have 4 posters on optimization:
- In Search of Adam’s Secret Sauce
- Is your batch size the problem? Revisiting the Adam-SGD gap in language modeling
- On the Interaction of Noise, Compression Role, and Adaptivity under (L0,L1)-Smoothness
- Universal Dynamics of Warmup Stable Decay: Understanding WSD beyond Transformers
Robert M. Gower retweeted
kexue.fm/archives/10996
The latest progress in finding better Newton-Schulz iterations for the msign operator. It directly derives the theoretical optimal solution through the equioscillation theorem and greedy transformation.
Original paper: papers.cool/arxiv/2505.16932
Robert M. Gower retweeted
How can classical numerical analysis help train deep nets faster? Climb aboard the Polar Express to find out... arxiv.org/abs/2505.16932
joint with @davpersson @gowerrobert + Chris Musco
Robert M. Gower retweeted
I have an opening for a post-doc position: I am looking for smart people with a strong CV in optimization and/or online learning
All my ex post-docs (@kwangsungjun, @mingruiliuCS, and @emsaad_p) became assistant professors, I'd like to continue this trend 😉
Please share it!
Robert M. Gower retweeted
Catch this poster at 3pm-5:30pm today:
I’ll be at ICLR next week presenting this paper co-written with Alex Damian (@alex_damian_). Would love to meet and chat about optimization in deep learning! My DMs are open - please reach out via DM or email.
openreview.net/forum?id=sIE2…
Robert M. Gower retweeted
arxiv.org/abs/2504.01898 looks pretty cool! model distillation setting seems to fit really well with what's needed to compute the polyak stepsize.
Robert M. Gower retweeted
Your optimization algorithm is a circuit!
In our NeurIPS 2024 Spotlight paper with Stephen Boyd, Tetiana Parshakova, Jaewook Suh @Jaewook_J_Suh, we present a framework for designing optimization algorithms using (electric) circuit theory.
(Video and paper links in reply)
(1/3)
Robert M. Gower retweeted
Go check out our #NeurIPS spotlight poster at 4:30pm today to hear about why heavy-tailed class imbalance is an important reason behind the success of Adam in training language models!
Paper: arxiv.org/abs/2402.19449, led by the great Fred Kunstner
If you are at NeurIPS, come see our poster Thursday at 4:30 on a reason Adam outperforms SGD on LLMs.
We can use the same setup to make problems where Adam outperforms SGD on CNNs, MLPs, or even linear models.
nips.cc/virtual/2024/poster/…
Robert M. Gower retweeted
Congratulations to my PhD student Kaja Gruntkowska for winning the 2024 Al-Kindi Research Award!
Robert M. Gower retweeted
Excited to present our latest paper on Variational Inference at NeurIPS'24, with @dianarycai, Chirag Modi, @gowerrobert, David Blei and Lawrence Saul from @FlatironInst!
Come see our spotlight poster Friday 7:30pm, East Exhibit Hall A-C #3900!
Fitting a multimodal target distribution with variational inference is hard. The fit and sampling often needs costly iterative methods.
Check out our spotlight NeurIPS paper where we use polynomials, score matching, and a direct solution for the fit. 1/x openreview.net/forum?id=ad97…
