Sílvia Casacuberta

We’re very excited to be organizing a 1-day workshop on “Recent advances in Regularity Lemmas (with and without calibration)” at Stanford on Tuesday, November 17! 
Program and details: regularitylemmas.github.io

In recent years, connections to multiaccuracy and multicalibration have led to a surge of work on regularity lemmas, with new results and applications in complexity theory, cryptography, learning theory, information theory, and communication complexity, among other areas.

We’re very excited to be organizing a 1-day workshop on “Recent advances in Regularity Lemmas (with and without calibration)” at Stanford on Tuesday, November 17! 
Program and details: regularitylemmas.github.io

It was great to attend @forcconf.bsky.social at Harvard this past week! I gave a talk on our paper “How Global Calibration Strengthens Multiaccuracy” (joint work with Parikshit Gopalan, Varun Kanade, and Omer Reingold), which we presented at FOCS this past December. arxiv.org/abs/2504.15206

...other multigroup fairness notions) has recently been shown to be powerful in other settings in TCS, including in omniprediction (arxiv.org/abs/2501.17205), in pseudoentropy characterizations (arxiv.org/abs/2507.05972), and in robust learning (arxiv.org/abs/2604.02555).

It was great to attend @forcconf.bsky.social at Harvard this past week! I gave a talk on our paper “How Global Calibration Strengthens Multiaccuracy” (joint work with Parikshit Gopalan, Varun Kanade, and Omer Reingold), which we presented at FOCS this past December. arxiv.org/abs/2504.15206

In this new paper, we show that we can obtain hardcore measures with optimal density from the weaker notion of calibrated multiaccuracy. In both the learning and complexity settings, our proofs demonstrate the complementary roles played by multiaccuracy and calibration.

It was great to attend @forcconf.bsky.social at Harvard this past week! I gave a talk on our paper “How Global Calibration Strengthens Multiaccuracy” (joint work with Parikshit Gopalan, Varun Kanade, and Omer Reingold), which we presented at FOCS this past December. arxiv.org/abs/2504.15206

A similar picture emerges in the setting of complexity theory. We know, from the work of Trevisan, Tulsiani, and Vadhan, that from the Regularity Lemma (= multiaccuracy theorem) we can prove Impagliazzo’s Hardcore Lemma, albeit with suboptimal density delta.

It was great to attend @forcconf.bsky.social at Harvard this past week! I gave a talk on our paper “How Global Calibration Strengthens Multiaccuracy” (joint work with Parikshit Gopalan, Varun Kanade, and Omer Reingold), which we presented at FOCS this past December. arxiv.org/abs/2504.15206

We show that for many classes C we can construct a predictor p that is perfectly C-multiaccurate, yet p is completely uncorrelated with the labels (even though some concepts in C do have high correlation with the labels).

It was great to attend @forcconf.bsky.social at Harvard this past week! I gave a talk on our paper “How Global Calibration Strengthens Multiaccuracy” (joint work with Parikshit Gopalan, Varun Kanade, and Omer Reingold), which we presented at FOCS this past December. arxiv.org/abs/2504.15206

Excited to be attending #ICLR in Rio this week! I will be presenting our paper (joint with Moritz Hardt) on the sample complexity of treatment allocation. I will be at Pavilion 4 on Saturday 3:15-5:45 pm (poster session 4, #4209). iclr.cc/virtual/2026...

Hello Bluesky! Joining the app in time to say that I’ll be attending NeurIPS this week! I will be presenting our paper (joint work with Varun Kanade) on building predictors that abstain optimally and fairly. neurips.cc/virtual/2025...