Completed
Neural networks and approximation theory
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Neural Networks as Interacting Particle Systems
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Machine learning and neural networks
- 3 3-spin model on the high-dimensional sphere
- 4 Neural networks and approximation theory
- 5 Functional formulation in the limit of large n
- 6 Parameters as particles with loss function as interacting potential
- 7 Error scaling - Central Limit Theorem (CLT)
- 8 Discrete training set and stochastic gradient descent
- 9 Limiting stochastic differential equation for SGD
- 10 Dean's equation for particles with correlated noise
- 11 Learning with Gaussian kemels
- 12 Learning with single layer networks with sigmoid nonlinearity
