Michael Roizner·Apr 9, 2025Counters, Filters, and HashingI’ve already written about counters and how to remove bias from them, but I haven’t touched on some important technical aspects…
Michael Roizner·Oct 6, 2024GPU-Powered Retrieval: Recent TrendNot too long ago, I learned about a new trend in our industry. It’s happening in an area where everything seemed to be working well…
Michael Roizner·Feb 15, 2024The Ultimate Question of Diversity, Exploration, and Value in Recommender SystemsReformulating the challenges of ‘beyond accuracy’ aspects through the lens of value optimization
Michael Roizner·Dec 26, 2023Personalization and Popularity BiasIs popularity bias just a convenient excuse?
InTDS ArchivebyMichael Roizner·Dec 6, 2023The Principled Approach to Early Ranking StagesA systematic method for designing and evaluating candidate generation and early ranking stages in recommender systems.A response icon1A response icon1
Michael Roizner·Nov 29, 2023Beyond External Embeddings: Integrating User Histories for Enhanced RecommendationsSome time ago, I wrote about how to use external embeddings. Today, let’s discuss a related (even overlapping) topic: how to use external…
InTDS ArchivebyMichael Roizner·Nov 24, 2023Two-Tower Networks and Negative Sampling in Recommender SystemsUnderstand the key elements that power advanced recommendation enginesA response icon4A response icon4
Michael Roizner·Nov 16, 2023Rating Systems: The Math Behind StarsMany services feature ratings of objects: products, movies, applications, organizations on a map. How should these be accurately…
InTDS ArchivebyMichael Roizner·Sep 23, 2023From Hacks to Harmony: Structuring Product Rules in RecommendationsDon’t let heuristics undermine your ML, learn to combine them
