← Search

Amir Joudaki

6 accepted papers

2026

Barriers for Learning in an Evolving World: Mathematical Understanding of Loss of Plasticity

ICLR 2026poster

Deep learning models excel in stationary settings but suffer from loss of plasticity (LoP) in non-stationary environments. While prior literature characterizes LoP through symptoms like rank collapse of representations, it often lacks a mechanistic explanation for why gradient descent fails to recov…

Cited by 0SourcecodeScholar
2025

Emergence of Globally Attracting Fixed Points in Deep Neural Networks With Nonlinear Activations

AISTATS 2025poster

Understanding how neural networks transform input data across layers is fundamental to unraveling their learning and generalization capabilities. Although prior work has used insights from kernel methods to study neural networks, a global analysis of how the similarity between hidden representations…

Cited by 0SourcecodeScholar
2024

Towards Training Without Depth Limits: Batch Normalization Without Gradient Explosion

ICLR 2024poster

Normalization layers are one of the key building blocks for deep neural networks. Several theoretical studies have shown that batch normalization improves the signal propagation, by avoiding the representations from becoming collinear across the layers. However, results on mean-field theory of batch…

2023

On Bridging the Gap between Mean Field and Finite Width Deep Random Multilayer Perceptron with Batch Normalization

ICML 2023poster

Mean-field theory is widely used in theoretical studies of neural networks. In this paper, we analyze the role of depth in the concentration of mean-field predictions for Gram matrices of hidden representations in deep multilayer perceptron (MLP) with batch normalization (BN) at initialization. It i…

Cited by 4SourcePDFScholar
2023

On the impact of activation and normalization in obtaining isometric embeddings at initialization

NeurIPS 2023poster

In this paper, we explore the structure of the penultimate Gram matrix in deep neural networks, which contains the pairwise inner products of outputs corresponding to a batch of inputs. In several architectures it has been observed that this Gram matrix becomes degenerate with depth at initializatio…

2021

Batch Normalization Orthogonalizes Representations in Deep Random Networks

NeurIPS 2021spotlight

This paper underlines an elegant property of batch-normalization (BN): Successive batch normalizations with random linear updates make samples increasingly orthogonal. We establish a non-asymptotic characterization of the interplay between depth, width, and the orthogonality of deep representations.…