Glossary term
Glossary term
Memory and Retrieval
An algorithm for minimizing the objective function during matrix factorization in recommendation systems, which allows a downweighting of the missing examples. WALS minimizes the weighted squared error between the original matrix and the reconstruction by alternating between fixing the row factorization and column factorization. Each of these optimizations can be solved by least squares convex optimization. For details, see the Recommendation Systems course.
Created for this library
A recommendation team uses Weighted Alternating Least Squares to learn user and item factors from implicit feedback data efficiently.
A music platform uses WALS for matrix factorization on its listening data where implicit signals dominate.
An e-commerce team uses WALS to factorize a sparse purchase matrix into user and item embeddings for its candidate generator.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License