Anna Győrffy-Kerekes

PhD student, ETHZ - ETH Zurich

Joined

September 2021

Names

Anna Győrffy-Kerekes (Preferred)

,

Anna Kerekes

Emails

****@gmail.com (Confirmed)

,

****@cam.ac.uk (Confirmed)

,

****@ethz.ch (Confirmed)

Personal Links

Career & Education History

PhD student

ETHZ - ETH Zurich (ethz.ch)

2023 – 2027

Undergrad student

University of Cambridge (cam.ac.uk)

2019 – 2023

Advisors, Relations & Conflicts

PhD Advisor

2023 – Present

PhD Advisor

Bernhard Schölkopf

2023 – Present

Expertise

No areas of expertise listed

Publications

To smooth a cloud or to pin it down: Expressiveness guarantees and insights on score matching in denoising diffusion models
Teodora Reu, Francisco Vargas, Anna Kerekes, Michael M. Bronstein
- UAI 2024 poster
- Readers: Everyone
Lessons from Identifiability for Understanding Large Language Models
Patrik Reizinger, Szilvia Ujváry, Anna Mészáros, Anna Kerekes, Wieland Brendel, Ferenc Huszár
- ICLR 2024 Workshop MEFoMo Withdrawn Submission
- Readers: Everyone
Position: Understanding LLMs Requires More Than Statistical Generalization
Patrik Reizinger, Szilvia Ujváry, Anna Mészáros, Anna Kerekes, Wieland Brendel, Ferenc Huszár
- ICML 2024 Spotlight
- Readers: Everyone
Expressiveness Remarks for Denoising Diffusion Based Sampling
Francisco Vargas, Teodora Reu, Anna Kerekes
- Published: 20 Jun 2023, Last Modified: 18 Jul 2023
- AABI 2023
- Readers: Everyone
Rethinking Sharpness-Aware Minimization as Variational Inference
Szilvia Ujváry, Zsigmond Telek, Anna Kerekes, Anna Mészáros, Ferenc Huszár
- Published: 23 Nov 2022, Last Modified: 05 May 2023
- OPT 2022 Poster
- Readers: Everyone
Depth Without the Magic: Inductive Bias of Natural Gradient Descent
Anna Kerekes, Anna Mészáros, Ferenc Huszár
- Published: 28 Jan 2022, Last Modified: 13 Feb 2023
- ICLR 2022 Submitted
- Readers: Everyone

Co-Authors