Learning To Collide: Recommendation System Model Compression With Learned Hash Functions

Abstract

A key characteristic of deep recommendation models is the immense memory requirements of their embedding tables. These embedding tables can often reach hundreds of gigabytes which increases hardware requirements and training cost. A common technique to reduce model size is to hash all of the categorical variable identifiers (ids) into a smaller space. This hashing reduces the number of unique representations that must be stored in the embedding table; thus decreasing its size. However, this approach introduces collisions between semantically dissimilar ids that degrade model quality. We introduce an alternative approach, Learned Hash Functions, which instead learns a new mapping function that encourages collisions between semantically similar ids. We derive this learned mapping from historical data and embedding access patterns. We experiment with this technique on a production model and find that a mapping informed by the combination of access frequency and a learned low dimension emb

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyghaemmaghami2022learning

Related papers

• Multi-probe Zero Collision Hash (MPZCH): Mitigating Embedding Collisions And Enhancing Model Freshness In Large-scale Recommenders (2026)0.00
• Learning To Hash For Recommendation: A Survey (2024)0.00
• Graphhash: Graph Clustering Enables Parameter Efficiency In Recommender Systems (2024)4.77
• Mem-rec: Memory Efficient Recommendation System Using Alternative Representation (2023)0.00
• Representation Learning For Efficient And Effective Similarity Search And Recommendation (2021)0.00
• Learning Similarity Preserving Binary Codes For Recommender Systems (2022)0.00
• Learning To Hash With Graph Neural Networks For Recommender Systems (2020)14.02
• Better Generalization With Semantic Ids: A Case Study In Ranking For Recommendations (2023)10.35