Collaborative Filtering

Oh my god it has been so long. This is where I talked about the Cosine Similarity Score, or maybe I just learned this in Linear Algebra.

https://developers.google.com/machine-learning/recommendation/collaborative/matrix

It seems however, that collaborative filtering is outdated, and I saw this hot new startup backed by Y-Combinator by a guy who used to do YouTube? who want to provide modern recommender systems to many companies, who don’t have the resources to pay millions.

CS341: teacher talked about this when motivating how to count inversions. If $i<j$, you count the number of times $A[i]>A[j]$, and that somehow helps with defining similarities…?

Concepts

• Matrix Factorization