Research map / Probabilistic and scientific modeling / Dynamical systems
The balance between feature learning and collapse in generative dynamical systems
Julian Brandon, Bruno Loureiro, N Alex Cayco Gajic, Arthur Pellegrino
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Theoretical study of how implicit biases in learning dynamics of generative models based on dynamical systems (normalising flows, diffusion) balance feature learning against collapse.
Problem
Representation learning dynamics in generative models built on dynamical systems, such as normalising flows and diffusion models, are poorly understood compared with deep learning theory for discriminative networks.
Approach
Analyses implicit biases of learning dynamics in generative dynamical-system models to characterise when feature learning occurs versus when representations collapse; the abstract is truncated so the specific method is not stated.
Closest papers by meaning
- A solvable model of learning generative diffusion: theory and insights (NeurIPS 2025 · similarity 0.89)
- Why Diffusion Models Don’t Memorize: The Role of Implicit Dynamical Regularization in Training (NeurIPS 2025 · similarity 0.88)
- Contrast encodes inductive bias: separating slow noise from dynamics in predictive representation learning (NeurIPS 2026 · similarity 0.87)
- When and how can inexact generative models still sample from the data manifold? (NeurIPS 2025 · similarity 0.87)
- Gradient Variance Reveals Failure Modes in Flow-Based Generative Models (NeurIPS 2025 · similarity 0.87)
- A Closer Look at Model Collapse: From a Generalization-to-Memorization Perspective (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
