Zahri-Edu › Math-University › Mathematics for Machine Learning & AI

AI 141

Differential Equations & Dynamical Systems in AI

Neural ODEs, gradient flows, diffusion and score-based generative models via SDEs, and physics-informed neural networks.

In preparation

The lecture notes for this course are being written. Below is the planned table of contents; chapters will be published one by one.

Table of contents

5 chapters · 12 sections

1

Dynamical Systems

  1. 1.1Flows and stability
  2. 1.2Gradient flows
2

Neural ODEs

  1. 2.1Residual networks as Euler steps
  2. 2.2Adjoint method
  3. 2.3Continuous normalising flows
3

Diffusion Models

  1. 3.1Forward noising as an SDE
  2. 3.2Score functions
  3. 3.3Reverse-time SDE and sampling
4

Physics-Informed Neural Networks

  1. 4.1PDE residual losses
  2. 4.2Examples: heat and Burgers equations
5

Multi-Agent Learning

  1. 5.1Collective dynamics
  2. 5.2Swarm intelligence