MLE @Waymo | Ph.D. @USCViterbi (@icaroslab) | cult leader of @pyribs
City of Angels, California
Joined July 2019
- Tweets575
- Following507
- Followers323
- Likes1.1K
New histogram and CDF visualizations now available in @pyribs 0.12.0. Go check it out! docs.pyribs.org/en/stable/wh…
PS: The inspiration for these plots came from the seminal CVT archives work by @v_vassiliades and @jb_mouret 🙂 arxiv.org/abs/1610.05729
This release also features new support for threshold decay in Novelty Search, thanks to Alejandro Marrero!
github.com/icaros-usc/pyribs…
As always, reach out with any questions, comments, or diatribes💜 discord.gg/QxhcJSqZ8G
Just released @pyribs 0.11.0! Notable changes include the addition of Novelty Search with Local Competition and DDS-CNF by @efsiatras, and a bug fix by @zobarPk. As always, more details are available at docs.pyribs.org/en/stable/wh… Cheers!
Excited to announce the release of @pyribs 0.10.0! Pyribs 0.10.0 introduces new algorithms (Discount Model Search and Dominated Novelty Search) and several features intended to make the library more flexible. Here are the highlights🧵
Furthermore, we have made a number of breaking changes to the CVTArchive, in particular so that it is easier to specify centroids and distance methods. We wrote a new tutorial that covers some of these changes: docs.pyribs.org/en/stable/tu…
And that's a wrap! For a more detailed look into this release, see our What's New page: docs.pyribs.org/en/stable/wh… And of course, reach out on GitHub, Twitter, or Discord (discord.gg/QxhcJSqZ8G) if you have any questions!
Ecstatic to share our #ICLR2026 Oral paper, “Discount Model Search for Quality Diversity Optimization in High-Dimensional Measure Spaces”, in collaboration with @xuster12 @tehqin17 @snikolaidis19
Website: discount-models.github.io/
Presentation @iclr_conf: iclr.cc/virtual/2026/poster/…
Problems like the one above are specified by providing a dataset of images/measures. Given the ubiquity of datasets in machine learning, we are excited to see what problems can be framed in a similar way and tackled with DMS.
To learn more, feel free to reach out! Or catch us at ICLR at Poster Session 5 (Saturday 4/25, 10:30AM - 1:00PM) and Oral Session 6C (Saturday 4/25, 3:15PM - 4:45PM).
Website: discount-models.github.io/
Presentation @iclr_conf: iclr.cc/virtual/2026/poster/…
