PeopleofSuperintelligence

Researcher

Nadav Cohen

Also known as נדב כהן

Nadav Cohen is an associate professor of computer science and AI at Tel Aviv University and the co-founder and CTO of Imubit, working on the mathematical foundations of deep learning and its use in physical AI. After a dual Technion degree in electrical engineering and mathematics, he earned a PhD under Amnon Shashua at the Hebrew University of Jerusalem, and a 2016 COLT paper with Or Sharir and Shashua used tensor analysis to prove an exponential size advantage for deep networks over shallow ones. As a postdoctoral fellow at the Institute for Advanced Study in Princeton from 2017, he worked with Sanjeev Arora and Elad Hazan, and their 2018 ICML paper showed that increasing depth can sometimes speed up optimization. He joined Tel Aviv University in 2019 and became an ELLIS Scholar in 2020, and Imubit, which he co-founded in 2016, announced $50 million in total funding in 2021. An ERC Starting Grant followed in 2024, and the US National Academy of Inventors elected him a Senior Member in 2026.

Now at
Tel Aviv University and Imubit
Born
1984
Badges

Years link to their sources.

Timeline

  1. By 2015

    A PhD with Amnon Shashua

    He earns a dual BSc in electrical engineering and mathematics, both summa cum laude, in the Technion Excellence Program, and a direct-track PhD in the School of Computer Science and Engineering at the Hebrew University of Jerusalem under Amnon Shashua. In 2015, he receives a Google PhD Fellowship in machine learning.

  2. 2015–2016

    Tensor analysis of deep networks

    In September 2015, he submits 'On the Expressive Power of Deep Learning: A Tensor Analysis' to arXiv with Or Sharir and Amnon Shashua, and it appears at COLT in 2016. The paper ties a shallow network to CP decomposition and a deep network to Hierarchical Tucker decomposition, and proves that, apart from a negligible set, every function a polynomial-size deep network implements needs exponential size in a shallow one.

  3. 2017–2018

    Institute for Advanced Study postdoc

    He joins the Institute for Advanced Study in Princeton as a postdoctoral fellow in September 2017, in the School of Mathematics and hosted by Sanjeev Arora and Elad Hazan. He is also a Zuckerman Israeli Postdoctoral Scholar in the 2017–2018 cohort, with Hazan as supervisor in Princeton's computer science department. With Arora and Hazan, he publishes 'On the Optimization of Deep Networks: Implicit Acceleration by Overparameterization' at ICML in 2018, which shows that increasing depth can sometimes speed up optimization.

  4. 2019

    Joining Tel Aviv University

    He joins the School of Computer Science & AI at Tel Aviv University as a faculty member in July 2019. His NeurIPS 2019 paper 'Implicit Regularization in Deep Matrix Factorization', with Sanjeev Arora, Wei Hu and Yuping Luo, finds that adding depth to a matrix factorization strengthens gradient descent's tendency toward low-rank solutions.

  5. 2020–2021

    ELLIS Scholar and Google award

    He is elected an ELLIS Scholar in December 2020. In 2021, he receives a Google Research Scholar Award for a project titled 'A Dynamical Theory of Deep Learning'.

  6. August 2021

    Imubit comes out of stealth

    Imubit, the company he co-founded in 2016 with Gil Cohen and Meir Tseitlin, announces a $30 million Series B round in August 2021 that brings its total funding to $50 million. The company provides AI-based process optimization for refiners and chemical operators, and he serves as its Chief Scientist.

  7. 2024–2026

    ERC grant, tenure and honors

    He receives an ERC Starting Grant in September 2024 and is promoted to associate professor with tenure in December 2024. In 2025, he coauthors a NeurIPS paper with Yonatan Slutzky, Yotam Alexander and Noam Razin that proves the implicit bias of structured state space models can be poisoned with clean labels. In February 2026, he is elected a Senior Member of the US National Academy of Inventors; he receives the Israel AI Safety Research Prize in May 2026 and appears on the TLV AI 50 list in September 2026.

  8. As of October 2026

    Associate professor and Imubit CTO

    He is an associate professor of computer science and AI at Tel Aviv University and the co-founder, CTO and President of Imubit. His academic group develops mathematical theories and systematic experiments aimed at principled deep learning algorithms, and his industry labs develop world models for decision making in physical AI, with systems in production at global scale.

Career

  1. Since 2019Tel Aviv UniversityFaculty Member
  2. Since 2016ImubitCo-founder, Chief Scientist
  3. 2017–2018Institute for Advanced StudyPostdoctoral Fellow

Education

  1. TechnionBSc in electrical engineering and mathematics
  2. Hebrew University of JerusalemPhD

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 Tel Aviv University. 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. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Tel Aviv University. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  3. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Tel Aviv University. 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.
  4. Google has the highest combined h-index among AI organizationsIncludes Tel Aviv University. 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.

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 Nadav Cohen yet. Send us a free-licensed photo.

  1. Nadav Cohen — Biography cohennadav.com. Checked 21 October 2026.
  2. Announcing Google’s 2015 Global PhD Fellows research.google. Checked 21 October 2026.
  3. On the Expressive Power of Deep Learning: A Tensor Analysis arxiv.org. Checked 21 October 2026.
  4. Nadav Cohen cohennadav.com. Checked 21 October 2026.
  5. Dr. Nadav Cohen | Zuckerman STEM Leadership Program zuckermanstem.org. Checked 21 October 2026.
  6. On the Optimization of Deep Networks: Implicit Acceleration by Overparameterization arxiv.org. Checked 21 October 2026.
  7. Implicit Regularization in Deep Matrix Factorization proceedings.neurips.cc. Checked 21 October 2026.
  8. Announcing the 2021 Research Scholar Program Recipients research.google. Checked 21 October 2026.
  9. Imubit announces arrival with $50 million in total funding for AI process optimization platform calcalistech.com. Checked 21 October 2026.
  10. The Implicit Bias of Structured State Space Models Can Be Poisoned With Clean Labels arxiv.org. Checked 21 October 2026.
  11. TLV AI 50 — The 50 Israelis Expanding the Frontier of AI 2026.tlvai50.com. Checked 21 October 2026.
  12. Leadership imubit.com. Checked 21 October 2026.