Daniil Dmitriev

I am a postdoc at the University of Pennsylvania, hosted by Yuting Wei in the Department of Statistics and Data Science at the Wharton School.

I work on theoretical foundations of machine learning, with a focus on high-dimensional statistics and modern generative models.

I completed my PhD at ETH Zurich, advised by Afonso Bandeira and Fanny Yang, and supported by the ETH AI Center and ETH FDS initiative.

Google Scholar / CV / LinkedIn / X / Blue sky

Contact me at: daniildmitriev32@gmail.com.

Selected publications and preprints

Theory of discrete diffusion models

Provably adaptive sampling with uniform and remasking discrete diffusion models

arxiv preprint, 2026

DD, Zhihan Huang, Yuting Wei

Efficient sampling with discrete diffusion models: sharp and adaptive guarantees

arxiv COLT, 2026

DD*, Zhihan Huang*, Yuting Wei (* here and below denotes equal contribution)

Learning theory and robust statistics

Learning in an echo chamber: online learning with replay adversary

arxiv SODA, 2026

DD, Harald Eskelund Franck, Carolin Heinzler, Amartya Sanyal (αβ order)

Robust mixture learning when outliers overwhelm small groups

arxiv / poster NeurIPS, 2024

DD*, Rares-Darius Buhai*, Stefan Tiegel, Alexander Wolters, Gleb Novikov, Amartya Sanyal, David Steurer, Fanny Yang

On the growth of mistakes in differentially private online learning: a lower bound perspective

arxiv / poster COLT, 2024

DD, Kristof Szabo, Amartya Sanyal

High-dimensional probability

The Lovász number of random circulant graphs

arxiv SampTA, 2025

Afonso S. Bandeira, Jarosław Błasiok, DD, Ulysse Faure, Anastasia Kireeva, Dmitriy Kunisky (αβ order)

Asymptotics of learning with deep structured (random) features

arxiv / poster ICML, 2024

Dominik Schröder*, DD*, Hugo Cui*, Bruno Loureiro

Greedy heuristics and linear relaxations for the random hitting set problem

arxiv APPROX, 2024

Gabriel Arpino, DD, Nicolo Grometto (αβ order)

Deterministic equivalent and error universality of deep random features learning

arxiv / video / poster ICML, 2023

Dominik Schröder, Hugo Cui, DD, Bruno Loureiro

Invited and contributed talks

Learning-driven Algorithms and Machine-aided Proofs (LAMP)

TTIC, 2026

IMS Annual Meeting

Salzburg, 2026

Foundations of Responsible Computing (FORC)

video Harvard, 2026

Youth in High Dimensions

video ICTP, 2025

DACO seminar

ETH Zurich, 2024

Youth in High Dimensions

video ICTP, 2024

Delta seminar

University of Copenhagen, 2024

Graduate seminar in probability

ETH Zurich, 2023

Workshop on Spin Glasses

video Les Diablerets, 2022

Teaching and Service

ETH Zurich, TA: Mathematics of Data Science (Fall 2021), Mathematics of Machine Learning (Spring 2022)

EPFL, TA: Artificial Neural Networks (Spring 2020, Spring 2021)

Supervising MSc theses at ETH Zurich: Carolin Heinzler (Fall 2023), Krish Agrawal, Ulysse Faure (Spring 2024)

Reviewer (Conferences): NeurIPS 2024 (Top reviewer), ICML 2025, ICLR 2025, ICLR 2026, COLT 2025, COLT 2026

Reviewer (Journals): TMLR, SIAM Journal on Mathematics of Data Science

Design and source code from Leonid Keselman's website