eytan bakshy

Research director at Meta leading the Adaptive Experimentation team. Bayesian optimization, Bayesian ML, bandits, RL, experiments, causal inference.

[eytan bakshy @eytan]

Ax 1.0 includes simplified interfaces for efficient optimization and experiment understanding, https://t.co/dT9dPaI2aC, as well as APIs for aiding the development, integration, and benchmarking of novel methods for BayesOpt and active learning with BoTorch.

[eytan bakshy @eytan]

Proud to share the 1.0 release of Ax, our platform for adaptive experimentation platform. We've been using it for the past 6 years for optimizing everything from end-to-end AI systems, recommender systems, AR hardware, and material science applications.
https://t.co/19aLoJ6jJU

[eytan bakshy @eytan]

In our paper, we demonstrate superior optimization performance on tiny budgets for a variety of tasks, including single- and multi-objective, high-dimensional inputs, discrete inputs+noisy outcomes, including parallel/async and early termination settings. https://t.co/eaco3e5Avk

[Jihao Andreas Lin @JihaoAndreasLin]

Big thanks to my amazing collaborators: Sebastian Ament, @MaxBalandat , @davidmeriksson , @jmhernandez233 , and @eytan !

If you are attending ICML 2025, feel free to check out our poster and have a chat with us!
Thu 11:00am - 1:30pm in East Exhibition Hall A-B, # E-1308

[eytan bakshy @eytan]

This work was a collaboration between the Meta Adaptive Experimentation and Infrastructure Data Science teams, UIUC, and Amrize. This mix was designed using batch multi-objective Bayesian optimization with BoTorch. Source code and data is available at https://t.co/ueWl4wwlcZ

[eytan bakshy @eytan]

Did you know that concrete accounts for 8% of global CO2 emissions? I'm delighted to share that after 2 years of development, we've successfully deployed our strong, low-carbon emission "AI concrete mix" at the latest Meta datacenter.
https://t.co/sSQ9dDc8iB

[eytan bakshy @eytan]

@leica_camera is there a way to provide customer feedback? For https://t.co/rDu8Cawk2o it only provides contact for repairs or public relations. Q3's lack of exposure preview in EVF w/o shutter button makes it impossible to effectively use exposure compensation dial+histogram.

[eytan bakshy @eytan]

@XingyouSong @JPDuerholt kudos to @XingyouSong and team for keeping up a high bar for OSS BayesOpt packages and publishing such extensive evaluations!

[eytan bakshy @eytan]

@XingyouSong @JPDuerholt That said, there is a lot of room for improvement in making sure the team's latest research ends up in OSS Ax and is dispatched automatically. We've also been using much improved support for mixed spaces internally, and hope to get that out in the near future.

[eytan bakshy @eytan]

Doing research on probabilistic modeling, decision-making under uncertainty, or efficient learning and optimization? Come to NeurIPS to present your work! Submissions due August 29th. https://t.co/2wkNY6WLmB

[eytan bakshy @eytan]

Interested in AutoML, active learning, Bayesian machine learning, or Bayesian optimization? We are hiring post-docs for the Adaptive Experimentation team! Experience with BoTorch, GenAI, or causal inference is a plus! Position in NYC, SF, and Menlo Park.

https://t.co/qr4339ZFAJ

[eytan bakshy @eytan]

Update: the meta-reviews are now up!

[eytan bakshy @eytan]

Is there a bug in OpenReview for #UAI2024? We submitted 3 papers, and none of the reviewers updated their responses and there are only reject or accept decisions—no meta-review.

[eytan bakshy @eytan]

Curious if anybody else had this experience.

[Alexander Terenin @avt_im]

Software package announcement!

The Geometric Kernels package, which implements various geometric Gaussian process models from my papers - on manifolds, graphs, and similar - is now on PyPi!

It can be installed with `pip install geometric_kernels`.

Check it out!

[eytan bakshy @eytan]

@rvalenciaaz Super cool work, @rvalenciaaz and team! The github+readme are very instructive as well!