Intro to Neural Network Gaussian Processes

Introductory explanation of the surprising result that wide neural networks are equivalent to Gaussian processes, and why it matters.

=== References (also at the end of the video) ===
Lee et. al. (2018) 'Deep Neural Networks as Gaussian Processes' https://arxiv.org/abs/1711.00165v3

Jascha Sohl-Dickstein. 'The Wide limit of Neural Networks: NNGP and NTK' (Streamed 2021). Data ICMC Understanding Machine Learning, Lecture 2.    • Lecture 2: The Wide limit of Neural N...  

Matthews et al. 'Gaussian Process Behaviour in Wide Deep Neural Networks' (2018) https://arxiv.org/abs/1804.11271v2

Jospin et al. 'Hands-On Bayesian Neural Networks — A Tutorial for Deep Learning Users' in IEEE Computational Intelligence Magazine, vol. 17, no. 2, pp. 29-48, May 2022, doi: 10.1109/MCI.2022.3155327

Lee et al. 'Wide neural networks of any depth evolve as linear models under gradient descent' (2020) Journal of Statistical Mechanics: Theory and Experiment https://arxiv.org/abs/2002.08526

Hardt and Recht. 'Patterns, Predictions, and Actions: Foundations of Machine Learning' (2022) https://mlstory.org/index.html