Large-Scale Parallel Recommender System in Julia
summary
A demonstration of a movie recommender system implemented end-to-end in Julia that uses matrix factorization (ALS or a similar algorithm). It reports dataset size, measured training time and throughput, concrete parallel scaling across cores and nodes with speedup numbers, accuracy metrics (RMSE or precision@k), online inference latency, memory and network usage, and whether code or binaries reproduce the results.
AI-written summary · about
chapters (5)
- 0:00Parallel recommender systems
- 3:30Large-scale challenges
- 8:50Julia parallel processing
- 13:56Live demo
- 16:37Summary
related talks
01h 31mHands on - Build a Recommender system
58mRecommender Systems – A Walkthrough
38mBuilding Recommender System in PHP8
48mCollective Intelligence
57mHands-On with LLM-Powered Recommenders: Hybrid Architectures
37mA primer on recommendation systems.
29mFurther enhancements of recommender systems using deep learning
01h 38mBuilding Recommender Systems Using Python