AI Paper Atlas

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

Everyone's Preference Changes Differently: A Weighted Multi-Interest Model For Retrieval

Hui Shi, Yupeng Gu, Yitong Zhou, Bo Zhao, Sicun Gao, Jishen Zhao

ICML 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Recommendation representation learning

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
two-stage training (freeze cluster weights at 1, then unfreeze) instead of joint training (ablation)MIP · AmazonAUC: 0.576 -> 0.806Improved
two-stage training instead of joint training (ablation)MIP · TaobaoAUC: 0.803 -> 0.885Improved
two-stage training instead of joint training (ablation)MIP · MovieLensAUC: 0.924 -> 0.930Mixed
learned personalized cluster weights instead of equal weights or exponential decay (ablation)MIP · Amazon, Taobao, MovieLens, PinterestAUC / recallImproved
NLL loss instead of triplet loss (alpha 0.2, 0.5, 0.8) (ablation)MIP · MovieLensAUC: 0.885 -> 0.930Improved
add positional and temporal encoding to item embedding (ablation)MIP · AmazonAUC: 76.34 -> 79.31Improved
add positional or temporal encoding (ablation)MIP · TaobaoAUC: 82.06 -> 86.59Improved
add positional and temporal encoding (ablation)MIP · MovieLensAUC: 95.26 -> 94.61Hurt
add sinusoid temporal and positional encoding (ablation)MIP · PinterestAUC: 0.8923 -> 0.8926No clear effect
replace sinusoid temporal encoding with one-hot or two-hot (ablation)MIP · PinterestAUC: 0.8921 -> 0.8850Hurt
multi-interest MIP versus single user embedding TiSASRecMIP · PinterestAUC: 0.850 -> 0.893Improved
self-attention multi-interest (MIP) instead of global-query attention (ComiRec)MIP · PinterestAUC: 0.875 -> 0.893Improved

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