Opening Neural Networks and the Learning Function
Neural Networks and the Learning Function
- Length
- 56 min
- Level
- Undergraduate
- Recorded
- MIT2019
Chapter 1 of 7: Before the lecture
Recovers point positions from squared pairwise distances by factoring the dot product matrix .
Distance matrices record squared distances between points, and Strang shows how to recover the positions of the points from those distances alone. He first frames a neural network as a learning function of weights and feature vectors , trained by minimizing square, L1, hinge or cross-entropy loss. The main derivation writes each squared distance through inner products, splitting the distance matrix into rank-one terms and the dot product matrix . Finding then needs a matrix square root, from the eigenvalue factorization or the faster Cholesky factorization. Applications include wireless sensor networks, molecular conformation and manifold learning.
Distance matrices record squared distances between points, and Strang shows how to recover the positions of the points from those distances alone. He first frames a neural network as a learning function of weights and feature vectors , trained by minimizing square, L1, hinge or cross-entropy loss. The main derivation writes each squared distance through inner products, splitting the distance matrix into rank-one terms and the dot product matrix . Finding then needs a matrix square root, from the eigenvalue factorization or the faster Cholesky factorization. Applications include wireless sensor networks, molecular conformation and manifold learning.

This lecture
Lecture 31 of 34 of Matrix Methods in Data Analysis, Signal Processing, and Machine Learning.
Choose a node to see how it connects to this lecture. What comes before is on the leftat the top. What comes after is on the rightat the bottom. The ring is what it teaches. People and places are above. Other lectures and books are below.
Press a chosen node again to open it.
Your comment is kept while you sign up.Sign in