Alex Meterez

Alex Meterez — ameterez [at] g [dot] harvard [dot] edu

Hey, I'm Alex, and I am a CS PhD student at Harvard co-advised by Prof. Sham Kakade and Prof. Cengiz Pehlevan. I am fortunate to be supported by a Kempner Fellowship in my research. My main interests are in Deep Learning theory and neural network optimization.

Publications and preprints (*: equal co-first, †: equal co-last)

Code is available at the links.

In Submission

A Defense of the Quadratic Model

A. Meterez, P. A. Nair, D. Morwani, C. Pehlevan, S. Kakade, A. Damian

HiLD (Oral) @ ICML 2026 Compute Efficiency and Serial Runtime Tradeoffs for Stochastic Momentum Methods

D. Morwani*, A. Meterez*, P. A. Nair*, C. Pehlevan, S. Kakade

In Submission

The Recurrent Transformer: Greater Effective Depth and Efficient Decoding

C. Oncescu, D. Morwani, S. Jelassi, A. Meterez, M. Kwun, S. Kakade

HiLD (Oral) @ ICML 2026

Anytime Pretraining: Horizon-Free Learning-Rate Schedules with Weight Averaging

A. Meterez*, P. A. Nair*, D. Morwani*, C. Pehlevan, S. Kakade

ICLR 2026

Seesaw: Accelerating Training by Balancing Learning Rate and Batch Size Scheduling

A. Meterez*, D. Morwani*, J. Wu, C. Oncescu, C. Pehlevan, S. Kakade

OPT @ NeurIPS 2025 A Simplified Analysis of SGD for Linear Regression with Weight Averaging

A. Meterez*, D. Morwani*, C. Oncescu, J. Wu, C. Pehlevan, S. Kakade

COLM 2025

Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining

R. Zhao*, A. Meterez*, S. Kakade, C. Pehlevan, S. Jelassi†, E. Malach†

ICLR 2025

The Optimization Landscape of SGD Across the Feature Learning Strength

A. Atanasov*, A. Meterez*, J. Simon*, C. Pehlevan

NeurIPS 2024

Why do Learning Rates Transfer? Reconciling Optimization and Scaling Limits for Deep Learning

L. Noci*, A. Meterez*, T. Hofmann, A. Orvieto

ICLR 2024

Towards Training Without Depth Limits: Batch Normalization Without Gradient Explosion

A. Meterez*, A. Joudaki*, F. Orabona, A. Immer, G. Rätsch, H. Daneshmand

RECOMB 2023

Aligning distant sequences to graphs using long seed sketches

A. Joudaki*, A. Meterez*, H. Mustafa, R. Groot Koerkamp, A. Kahles, G. Rätsch