Sign inSign in
ConceptAI & ML

Diffusion models

Generative models that add random noise to data in a forward process and learn a reverse denoising process to generate new samples.
as Russ Tedrake defines itMITThe wording closest to all 7 definitions
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Opens up

Concepts that name this one as a prerequisite.

Lecturers

7

Each in their own terms, from the easiest level.

  • Arthur Mattuck

    MIT · Undergraduate

    A mathematical model using a first-order differential equation analogous to the conduction model, describing how concentration changes as a substance diffuses across a semipermeable membrane.
    • 16:21First-order Linear ODEs: Steady-state and Transient Solutions
  • Phillip Isola

    MIT · Undergraduate

    A type of generative model that generates data by reversing a progressive noising process.
    • 3:16Generative Models: Basics
  • Sergey Levine

    UC Berkeley · Undergraduate

    Generative models that construct a fake training set by adding noise to data and then training a neural network to remove that noise.
    • 20:32Imitation Learning III: Inverse RL & Generative Adversarial Imitation Learning (GAIL)
  • Jim Harris

    National Capital Area Skeptics (NCAS) · Undergraduate

    A generative model that creates images from text instructions handed off by an LLM.
    • 41:47Mythbusting AI: Understanding Large Language Models and Their Limitations
  • Sayak Paul

    Stanford University · Graduate

    Generative models that iteratively denoise a random noise vector over a period of time to become a realistic image.
    • 3:35Transformers in Diffusion Models for Image Generation and Beyond
  • Anders Malthe-Sørenssen

    UT Austin · Graduate

    A generative model used to predict configurations of structures by reversing a noise process.
    • 35:03Physical Insights from Discriminative and Generative Models for Frictional Surface Design

Lectures that assume it

5

Connections

21

The lectures that define it or assume it, and the concepts next to it.

7 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.