Ouail Kitouni

4 papers · 2022–2024 · 3 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 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (10)

Venues

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

Top co-authors

Niklas Nolte (3) Mike Williams (3) Max Tegmark (1) Mike Rabbat (1) Niklas S Nolte (1) Adina Williams (1) Eric Michaud (1) Mark Ibrahim (1) Sokratis Trifinopoulos (1) Diane Bouchacourt (1)

Research topics

Machine Learning (3) Artificial Intelligence (2) Deep Learning (2) Core AI (2) Optimization & Theory · Machine Learning (2) Natural Language Processing (1) Application Areas (1) Applications (1) Architectures (1) Core Methods (1) Learning Types (1) Optimization & Theory · Deep Learning (1)

Keywords

representation learning (1) interpretability (1) information retrieval (1) inductive bia (1) lipschitz constant (1) next-token prediction (1) mechanistic interpretability (1) reversal curse (1) monotonic neural network (1) low-dimensional representation (1) neural network (1) phase diagram (1) grokking phenomenon (1) effective theory (1) generalization dynamic (1) knowledge storage (1) factorization curse (1) human-derived domain knowledge (1) nuclear physics concepts (1)

Papers

From Neurons to Neutrons: A Case Study in Interpretability ICML 2024 The Factorization Curse: Which Tokens You Predict Underlie the Reversal Curse and More NeurIPS 2024 Expressive Monotonic Neural Networks ICLR 2023 Towards Understanding Grokking: An Effective Theory of Representation Learning NeurIPS 2022