Amir Joudaki

5 papers · 2021–2025 · 4 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🗺️ Taxonomy Completionist (10) 🏆 Keyword Champion (2)

Venues

NeurIPS (2) AISTATS (1) ICLR (1) ICML (1)

Top co-authors

Hadi Daneshmand (4) Francis R. Bach (2) Gunnar Rätsch (1) Francesco Orabona (1) Thomas Hofmann (1) Francis Bach (1) Alexander Immer (1) Alexandru Meterez (1)

Research topics

Deep Learning (5) Machine Learning (4) Techniques (4) Optimization & Theory · Deep Learning (3) Optimization & Theory · Machine Learning (3) Architectures (1) Core Methods (1)

Keywords

batch normalization (3) mean field theory (2) gram matrix (2) neural network theory (1) signal propagation (1) implicit bia (1) concentration bound (1) activation function (1) layer normalization (1) orthogonal representation (1) network depth (1) gradient propagation (1) multi-layer perceptron (1) isometric embedding (1) neural network (1) gradient explosion (1) mean-field regime (1) gram matrice (1) hermite polynomial (1) rank collapse (1)

Papers

Emergence of Globally Attracting Fixed Points in Deep Neural Networks With Nonlinear Activations AISTATS 2025 Towards Training Without Depth Limits: Batch Normalization Without Gradient Explosion ICLR 2024 On Bridging the Gap between Mean Field and Finite Width Deep Random Multilayer Perceptron with Batch Normalization ICML 2023 On the impact of activation and normalization in obtaining isometric embeddings at initialization NeurIPS 2023 Batch Normalization Orthogonalizes Representations in Deep Random Networks NeurIPS 2021