Publications

This is a list of the publications I coauthored, including arxiv links.


Emmanuel Abbe, Elisabetta Cornacchia, Jan Hązła and Christopher Marquis (2022), "An initial alignment between neural network and target is needed for gradient descent to learn".
[arxiv]
Ido Nachum, Jan Hązła, Michael Gastpar and Anatoly Khina (2021), "A Johnson-Lindenstrauss Framework for Randomly Initialized CNNs".
[arxiv]
Elisabetta Cornacchia, Jan Hązła, Ido Nachum and Amir Yehudayoff (2021), "Regularization by Misclassification in ReLU Neural Networks".
[arxiv]
Jan Hązła, Alex Samorodnitsky and Ori Sberlo (2021), "On codes decoding a constant fraction of errors on the BSC". Symposium on Theory of Computing (STOC), pp. 1479-1488.
[link]
Elisabetta Cornacchia and Jan Hązła (2020), "Intransitive dice tournament is not quasirandom".
[arxiv]
Emmanuel Abbe, Jan Hązła and Ido Nachum (2021), "Almost-Reed-Muller codes achieve constant rates for random errors". IEEE Transactions on Information Theory, 67(12), pp. 8034-8050.
[arxiv]
Jan Hązła, Yan Jin, Elchanan Mossel and Govind Ramnarayan (2019), "A geometric model of opinion polarization".
[arxiv]
Jan Hązła (2020), "On arithmetic progressions in symmetric sets in finite field model". Electronic Journal of Combinatorics, 27(3), article no. P3.61.
[arxiv]
Jan Hązła, Ali Jadbabaie, Elchanan Mossel and M. Amin Rahimian (2019), "Reasoning in Bayesian opinion exchange networks is PSPACE-hard". Conference on Learning Theory (COLT), pp. 99:1614-1648.
[arxiv]
Jan Hązła, Elchanan Mossel, Nathan Ross and Guangqu Zheng (2020), "The probability of intransitivity in dice and close elections". Probability Theory and Related Fields, 178, pp. 99:951-1009.
[arxiv]
Jan Hązła, Ali Jadbabaie, Elchanan Mossel and M. Amin Rahimian (2021), "Bayesian decision making in groups is hard". Operations Research, 69(2), pp. 632-654.
[arxiv]
Jan Hązła (2016), "Same-set hitting with applications to parallel repetition". PhD thesis at ETH Zurich.
[link]
Jan Hązła, Thomas Holenstein and Anup Rao (2016), "Forbidden subgraph bounds for parallel repetition and the density Hales-Jewett theorem".
[arxiv]
Jan Hązła, Thomas Holenstein and Elchanan Mossel (2018), "Product space models of correlation: Between noise stability and additive combinatorics". Discrete Analysis, article no. 20.
[arxiv]
Conference version published as "Lower bounds on same-set inner product in correlated spaces". International Conference on Randomization and Computation (RANDOM) 2016, pp. 34:1-11.
Jan Hązła and Thomas Holenstein (2015), "Upper tail estimates with combinatorial proofs". Symposium on Theoretical Aspects of Computer Science (STACS), pp. 392-405.
[arxiv]