Binxu
  • Bio
  • Papers
  • Talks
  • News
  • Experience
  • Projects
  • Teaching
  • Recent & Upcoming Talks
    • IAIFI Summer Workshop Talk
    • Spring into Science Retreat Talk
    • Washington University Neuroscience Department Seminar
    • Yale Cognitive & Neural Computation Lab Talk
    • SfN Nanosymposium 2024
    • NAISYS2024 Conference Talk
    • KITP Deep Learning Program
    • SfN Nanosymposium 2023
    • PhD Thesis Defense
    • Kempner Institute Invited Lecture
    • Example Talk
  • Publications
    • An analytical theory of power law spectral bias in the learning dynamics of diffusion models
    • Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models
    • Factorized convolution models for predicting and interpreting neuronal tuning in natural images
    • Feature accentuation along the encoding axes of IT neurons uncovers hidden differences in model-brain alignment
    • On the importance of dynamic range and sample statistics in fitting neuronal encoding models
    • Parametric control along the encoding axes of IT neurons uncovers hidden differences in model-brain alignment
    • How do diffusion models learn and generalize on abstract rules for reasoning?
    • Neural Dynamics of Object Manifold Alignment in the Ventral Stream
    • Diverse capability and scaling of diffusion and auto-regressive models when learning abstract rules
    • Do diffusion models generalize on abstract rules for reasoning?
    • The Unreasonable Effectiveness of Gaussian Score Approximation for Diffusion Models and its Applications
    • Charting the Landscape of Ventral Stream Neural Code on Generative Image Manifolds
    • Diffusion models generate images like painters: an analytical theory of outline first, details later
    • The hidden linear structure in score-based models and its application
    • Understanding Learning Dynamics of Neural Representations via Feature Visualization at Scale
    • Factorized convolution models for interpreting neuron-guided images synthesis
    • High-performance Evolutionary Algorithms for Online Neuron Control
    • On the Level Sets and Invariance of Neural Tuning Landscapes
    • Tuning landscapes of the ventral stream
    • On the use of Cortical Magnification and Saccades as Biological Proxies for Data Augmentation
    • The Geometry of Deep Generative Image Models and its Applications
    • Visual prototypes in the ventral stream are attuned to complexity and gaze behavior
    • What constitutes understanding of ventral pathway function?
    • An example preprint / working paper
    • A mechanism for synaptic copy between neural circuits
    • Mutual information and information gating in synfire chains
    • An example journal article
    • An example conference paper
  • Projects
  • Blog
    • 🎉 Easily create your own simple yet highly customizable blog
    • 🧠 Sharpen your thinking with a second brain
    • 📈 Communicate your results effectively with the best data visualizations
    • 👩🏼‍🏫 Teach academic courses
    • ✅ Manage your projects
  • Projects
    • Pandas
    • PyTorch
    • scikit-learn
  • Experience
  • Teaching
    • Learn Python
    • Neuro140/240 - Biological and Artificial Intelligence
    • Neuro 120 - Introduction to Computational Neuroscience
    • Machine Learning from Scratch
    • BIOL 5648 - Coding and Statistical Thinking in the Neurosciences

An analytical theory of power law spectral bias in the learning dynamics of diffusion models

Jan 1, 2025·
Binxu Wang
· 0 min read
Cite
Type
Journal article
Publication
arXiv preprint arXiv:2503.03206
Last updated on Jan 1, 2025
Diffusion Science of AI
Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models Jan 1, 2025 →

© 2025 Me. This work is licensed under CC BY NC ND 4.0

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.