PeopleofSuperintelligence

Researcher

Idan Attias

Idan Attias is a machine learning theorist who studies the foundations of learning and data-driven sequential decision-making. He earned a PhD in computer science at Ben-Gurion University in 2024, supervised by Aryeh Kontorovich and Yishay Mansour. In 2024, he coauthored a paper on memorization and learning in stochastic convex optimization with Gintare Karolina Dziugaite, Mahdi Haghifam, Roi Livni and Daniel Roy, and ICML named it a Best Paper. His 2026 paper on the hardness of learning regular expressions, written with Lev Reyzin, Nathan Srebro and Gal Vardi, received an Elegance Paper Award at ALT 2026. He is a postdoctoral researcher at the IDEAL Institute in Chicago, hosted by Lev Reyzin, Avrim Blum and Nati Srebro, and will join the Technion's Computer Science Department as a faculty member in 2027.

Now at
IDEAL Institute and University of Illinois Chicago

Timeline

  1. 2019–2022

    Generalization bounds for robust learning

    With Aryeh Kontorovich and Yishay Mansour, he presents “Improved Generalization Bounds for Adversarially Robust Learning” at ALT 2019. The paper models a learner that must resist an adversary limited to a fixed number of possible corruptions for each input, and its journal version appears in JMLR in 2022.

  2. 2023

    Intern at the Vector Institute

    In August 2023, he starts as a research intern at the Vector Institute, working with Daniel Roy and Karolina Dziugaite.

  3. 2024

    A PhD at Ben-Gurion University

    He earns a PhD in computer science from Ben-Gurion University in 2024, supervised by Aryeh Kontorovich and by Yishay Mansour of Tel Aviv University and Google Research. His BSc in mathematics and computer science is from Tel Aviv University.

  4. 2024

    An ICML Best Paper

    In 2024, he coauthors “Information Complexity of Stochastic Convex Optimization: Applications to Generalization, Memorization, and Tracing” with Gintare Karolina Dziugaite, Mahdi Haghifam, Roi Livni and Daniel M. Roy. ICML 2024 names it a Best Paper, and the work studies how memorization and learning interact in stochastic convex optimization.

  5. By 2024

    Postdoc at the IDEAL Institute

    By September 2024, he is a postdoctoral researcher at the IDEAL Institute, hosted by Lev Reyzin (UIC) and Avrim Blum (TTIC), and works on the foundations of machine learning theory and data-driven sequential decision-making. In 2024, he is also selected as a Rising Star in Data Science (UCSD).

  6. 2025

    Traceability in convex optimization

    The 2025 paper “On Traceability in ℓp Stochastic Convex Optimization,” written with Sasha Voitovych, Mahdi Haghifam, Gintare Karolina Dziugaite, Roi Livni and Daniel M. Roy, shows a tradeoff between traceability and excess risk. A learning algorithm is traceable when analyzing its output can identify some of its training samples.

  7. 2026

    Elegance Paper Award at ALT

    His paper “On the Hardness of Learning Regular Expressions,” written with Lev Reyzin, Nathan Srebro and Gal Vardi, wins an Elegance Paper Award at ALT 2026. It studies the hardness of improperly learning regular expressions in the PAC model and with membership queries.

  8. As of October 2026

    Postdoc, then the Technion

    He is a postdoctoral researcher at the IDEAL Institute in Chicago and a postdoctoral research associate in UIC's Department of Mathematics, Statistics, and Computer Science, hosted by Lev Reyzin, Avrim Blum and Nati Srebro. He will join the Computer Science Department at the Technion as a faculty member in 2027.

Career

  1. NowIDEAL InstitutePostdoctoral Researcher
  2. NowUniversity of Illinois ChicagoPostdoctoral Research Associate
  3. From 2023Vector InstituteResearch Intern

Education

  1. 2024Ben-Gurion University of the NegevPhD
  2. Tel Aviv UniversityBSc

From the biography.

  1. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Idan Attias. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes University of Illinois Chicago. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes University of Illinois Chicago. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  4. 5 of the 25 people with the most-cited recent papers also lead on career citationsIncludes University of Illinois Chicago. We compare citations to 2023–2025 papers with career citations and h-index, after correcting reviewed publication-date errors: who leads, when they started and where they work.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Idan Attias yet. Send us a free-licensed photo.

  1. Idan Attias idanattias.com. Checked 21 October 2026.
  2. [1810.02180] Improved Generalization Bounds for Adversarially Robust Learning arxiv.org. Checked 21 October 2026.
  3. [Colloquium] 9/20 Talks at TTIC: Idan Attias, IDEAL Institute mailman.cs.uchicago.edu. Checked 21 October 2026.
  4. Postdoctoral Research Associate Idan Attias | Dept of Math, Stat, & Comp Sci | University of Illinois Chicago mscs.uic.edu. Checked 21 October 2026.
  5. ICML 2024 Awards icml.cc. Checked 21 October 2026.
  6. Information Complexity of Stochastic Convex Optimization: Applications to Generalization, Memorization, and Tracing proceedings.mlr.press. Checked 21 October 2026.
  7. TTIC Postdoctoral ttic.edu. Checked 21 October 2026.
  8. [Colloquium] TOMORROW: 12/17 Talks at TTIC: Idan Attias, IDEAL/UIC/TTIC Postdoc mailman.cs.uchicago.edu. Checked 21 October 2026.
  9. [2502.17384] On Traceability in ℓp Stochastic Convex Optimization arxiv.org. Checked 21 October 2026.
  10. On the Hardness of Learning Regular Expressions proceedings.mlr.press. Checked 21 October 2026.