AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems

Benjamin Coleman, Wang-Cheng Kang, Matthew Fahrbach, Ruoxi Wang, Lichan Hong, Ed H. Chi, Derek Zhiyuan Cheng

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add feature multiplexing (one shared table for all features) (ablation)Hashing trick · Criteo 2.5MBAUC: 79.44 -> 80.49Improved
add feature multiplexing (one shared table for all features) (ablation)Hashing trick · Movielens-1M 158kBAUC: 77.07 -> 82.00Improved
add feature multiplexing (one shared table for all features) (ablation)Hashing trick · Avazu 324kBAUC: 75.10 -> 76.86Improved
add feature multiplexing (one shared table for all features) (ablation)QR embedding · Criteo 2.5MBAUC: 79.74 -> 80.48Improved
add feature multiplexing (one shared table for all features) (ablation)PQ embedding · Criteo 2.5MBAUC: 79.82 -> 80.54Improved
add feature multiplexing (one shared table for all features) (ablation)PQ embedding · Avazu 324kBAUC: 76.48 -> 76.96Improved
add feature multiplexing (one shared table for all features) (ablation)PQ embedding · Movielens-1M 158kBAUC: 87.32 -> 88.09Improved
add feature multiplexing (one shared table for all features) (ablation)ROBE-Z embedding · Avazu 324kBAUC: 76.88 -> 76.94No clear effect
add feature multiplexing (one shared table for all features) (ablation)Hash embedding · Movielens-1M 1.6MBAUC: 88.28 -> 88.79Improved
add feature multiplexing (one shared table for all features) (ablation)Hash embedding · Criteo 2.5MBAUC: 80.40 -> 80.31Hurt
add feature multiplexing (one shared table for all features) (ablation)QR embedding · Criteo 25MBAUC: 80.29 -> 80.22Hurt
add feature multiplexing (one shared table for all features) (ablation)Hashing trick · Criteo training speed (25MB, CPU)steps/second: 29.2 -> 29.4No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy