Picture for Cengiz Pehlevan

Cengiz Pehlevan

Dynamically Learning to Integrate in Recurrent Neural Networks

Add code
Mar 24, 2025
Viaarxiv icon

Learning richness modulates equality reasoning in neural networks

Add code
Mar 12, 2025
Viaarxiv icon

Adaptive kernel predictors from feature-learning infinite limits of neural networks

Add code
Feb 11, 2025
Viaarxiv icon

Two-Point Deterministic Equivalence for Stochastic Gradient Dynamics in Linear Models

Add code
Feb 07, 2025
Viaarxiv icon

Deep Linear Network Training Dynamics from Random Initialization: Data, Width, Depth, and Hyperparameter Transfer

Add code
Feb 04, 2025
Viaarxiv icon

A precise asymptotic analysis of learning diffusion models: theory and insights

Add code
Jan 07, 2025
Viaarxiv icon

No Free Lunch From Random Feature Ensembles

Add code
Dec 06, 2024
Viaarxiv icon

Scaling Laws for Precision

Add code
Nov 07, 2024
Viaarxiv icon

Do Mice Grok? Glimpses of Hidden Progress During Overtraining in Sensory Cortex

Add code
Nov 05, 2024
Figure 1 for Do Mice Grok? Glimpses of Hidden Progress During Overtraining in Sensory Cortex
Figure 2 for Do Mice Grok? Glimpses of Hidden Progress During Overtraining in Sensory Cortex
Figure 3 for Do Mice Grok? Glimpses of Hidden Progress During Overtraining in Sensory Cortex
Figure 4 for Do Mice Grok? Glimpses of Hidden Progress During Overtraining in Sensory Cortex
Viaarxiv icon

A Brain-Inspired Regularizer for Adversarial Robustness

Add code
Oct 11, 2024
Figure 1 for A Brain-Inspired Regularizer for Adversarial Robustness
Figure 2 for A Brain-Inspired Regularizer for Adversarial Robustness
Figure 3 for A Brain-Inspired Regularizer for Adversarial Robustness
Figure 4 for A Brain-Inspired Regularizer for Adversarial Robustness
Viaarxiv icon