Search

What are you looking for?

  1. Learning Normalized Energy Models for Linear Inverse Problems

    Publication
    Generative diffusion models can provide powerful prior probability models for inverse problems in imaging, but existing implementations suffer from two key limitations: (i) the prior density is represented …

  2. There Will Be a Scientific Theory of Deep Learning

    Publication
    … learning mechanics should aspire to be its physics, mirroring the complementary relationship between biology and physics in the natural sciences. We also review and address common arguments …

  3. Learning a distance measure from the information-estimation geometry of data

    Publication
    We introduce the Information-Estimation Metric (IEM), a novel form of distance function derived from an underlying continuous probability density over a domain of signals. The IEM is …

  4. Learning normalized image densities via dual score matching

    Publication
    Learning probability models from data is at the heart of many machine learning endeavors, but is notoriously difficult due to the curse of dimensionality. We introduce a new …

  5. Classification-Denoising Networks

    Publication
    Image classification and denoising suffer from complementary issues of lack of robustness or partially ignoring conditioning information. We argue that they can be alleviated by unifying both tasks …

  6. A Rainbow in Deep Network Black Boxes

    Publication
    A central question in deep learning is to understand the functions learned by deep networks. What is their approximation class? Do the learned weights and representations depend on …

  7. On the universality of neural encodings in CNNs

    Publication
    We explore the universality of neural encodings in convolutional neural networks trained on image classification tasks. We develop a procedure to directly compare the learned weights rather than …

  8. Generalization in diffusion models arises from geometry-adaptive harmonic representations

    Publication
    We explore the universality of neural encodings in convolutional neural networks trained on image classification tasks. We develop a procedure to directly compare the learned weights rather than …

  9. Learning multi-scale local conditional probability models of images

    Publication
    Deep neural networks can learn powerful prior probability models for images, as evidenced by the high-quality generations obtained with recent score-based diffusion methods. But the means …