Ildus Sadrtdinov

According to our database, Ildus Sadrtdinov authored at least 6 papers between 2021 and 2026.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of five.

2026

Why Gaussian Diffusion Models Fail on Discrete Data?

Alexander Shabalin, Simon Elistratov, Viacheslav Meshchaninov, Ildus Sadrtdinov, Dmitry P. Vetrov

CoRR, April, 2026

2025

Can Training Dynamics of Scale-Invariant Neural Networks Be Explained by the Thermodynamics of an Ideal Gas?

Ildus Sadrtdinov, Ekaterina Lobacheva, Ivan Klimov, Mikhail I. Katsnelson, Dmitry P. Vetrov

CoRR, November, 2025

SGD as Free Energy Minimization: A Thermodynamic View on Neural Network Training.

Ildus Sadrtdinov, Ivan Klimov, Ekaterina Lobacheva, Dmitry P. Vetrov

CoRR, May, 2025

2024

Where Do Large Learning Rates Lead Us?

Ildus Sadrtdinov, Maxim Kodryan, Eduard Pokonechny, Ekaterina Lobacheva, Dmitry P. Vetrov

Proceedings of the Advances in Neural Information Processing Systems 37: Annual Conference on Neural Information Processing Systems 2024, 2024

2023

To Stay or Not to Stay in the Pre-train Basin: Insights on Ensembling in Transfer Learning.

Ildus Sadrtdinov, Dmitrii Pozdeev, Dmitry P. Vetrov, Ekaterina Lobacheva

Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

2021

On the Memorization Properties of Contrastive Learning.

Ildus Sadrtdinov, Nadezhda Chirkova, Ekaterina Lobacheva

CoRR, 2021