Eren C. Kızıldağ

e-mail :

I am a Distinguished Postdoctoral Fellow at the Department of Statistics at Columbia University. I received my PhD in Electrical Engineering and Computer Science (EECS) at the Massachusetts Institute of Technology (MIT), where I was very fortunate to be advised by Prof. David Gamarnik. At MIT, I was affiliated with the Laboratory for Information and Decision Systems (LIDS) and the Institute for Data, Systems, and Society (IDSS). You can find my CV here.

Research Interests

My research revolves around problems in high-dimensional statistics, theory of machine learning, and applied probability with an emphasis on computational aspects. In particular, I would like to understand the fundamental computational limits of such problems by studying the regimes of apparent hardness where efficient algorithms cease to exist. This is often guided by the insights provided by statistical physics and spin glass theory.



  1. Algorithms and Barriers in the Symmetric Binary Perceptron Model (Slides,Poster)
    IEEE Symposium on Foundations of Computer Science (FOCS), 2022 (To appear).
    with David Gamarnik, Will Perkins, and Changji Xu.
  2. Self-Regularity of Non-Negative Output Weights for Overparameterized Two-Layer Neural Networks (Slides)
    IEEE Transactions on Signal Processing.
    Conference version in IEEE International Symposium on Information Theory (ISIT), 2021
    with David Gamarnik and Ilias Zadik.
  3. Algorithmic Obstructions in the Random Number Partitioning Problem (Talk by David,Poster,Slides)
    Annals of Applied Probability (Major Revisions).
    Conference version in IEEE International Symposium on Information Theory (ISIT), 2022
    with David Gamarnik.
  4. Stationary Points of Shallow Neural Networks with Quadratic Activation Function (MIT ML Tea Talk,Updated Slides)
    with David Gamarnik and Ilias Zadik.
  5. Inference in High-Dimensional Linear Regression via Lattice Basis Reduction and Integer Relation Detection
    IEEE Transactions on Information Theory.
    with David Gamarnik and Ilias Zadik.
  6. Computing the Partition Function of the Sherrington-Kirkpatrick Model is Hard on Average (Talk,Slides)
    Annals of Applied Probability.
    Conference version in IEEE International Symposium on Information Theory (ISIT), 2020
    with David Gamarnik.
  7. High-Dimensional Linear Regression and Phase Retrieval via PSLQ Integer Relation Algorithm (Slides)
    IEEE International Symposium on Information Theory (ISIT), 2019.
    with David Gamarnik.
  8. Neural Networks and Polynomial Regression. Demystifying the Overparametrization Phenomena
    with Matt Emschwiller, David Gamarnik, and Ilias Zadik.

Other Biographical Notes

I earned my B.S. degree with highest honors (summa cum laude) in electrical and electronics engineering from Bogaziçi University, Istanbul, Turkey in 2014; and my M.S. degree in EECS from MIT in 2017. I did experimental research on magnetic resonance imaging (MRI) for my Master's thesis. Part of my thesis work appeared in the 24th Annual Meeting of International Society for Magnetic Resonance in Medicine (ISMRM), Singapore, 2016; and received a summa cum laude award (top 5% of all submitted works). Slides of my talk is available here.

I am a proud graduate of Ankara Science High School (AFL), class of 2010.