Research map / Label-efficient and robust learning / Recognition and long-tail
Dynamic Tensor Recommender Systems
Yanqing Zhang, Xuan Bi, Niansheng Tang, Annie Qu
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use DTRS instead of BPTF | tensor recommender · Simulation, independent errors, T2=8 | RMSE: 2.675 -> 1.570 | Improved |
| use DTRS instead of REM | tensor recommender · Simulation, independent errors, T2=8 | RMSE: 2.502 -> 1.570 | Improved |
| match working correlation to true independent errors (DTRSin vs DTRSar) (ablation) | DTRS · Simulation, independent errors, T2=8 | RMSE: 1.625 -> 1.570 | Improved |
| match working correlation to true AR-1 errors (DTRSar vs DTRSin) (ablation) | DTRS · Simulation, AR-1 errors, T2=12 | RMSE: 1.707 -> 1.632 | Improved |
| forecast further ahead (T2 from 8 to 12) | BPTF · Simulation, AR-1 errors | RMSE: 2.724 -> 3.181 | Hurt |
| forecast further ahead (T2 from 8 to 12) | DTRS · Simulation, independent errors | RMSE: 1.570 -> 1.660 | No clear effect |
| use DTRSin instead of BPTF | tensor recommender · IRI drug-store sales forecast 2011 | RMSE: 16.571 -> 12.504 | Improved |
| use DTRSin instead of REM | tensor recommender · IRI drug-store sales forecast 2011 | MAE: 4.432 -> 3.897 | Improved |
| use independent working correlation instead of AR-1 (ablation) | DTRS · IRI drug-store sales forecast 2011 | RMSE: 13.266 -> 12.504 | Improved |
Closest papers by meaning
- Smooth neighborhood recommender systems (JMLR 2019 · similarity 0.83)
- Scalable Demand-Aware Recommendation (NeurIPS 2017 · similarity 0.83)
- The Decoupled Extended Kalman Filter for Dynamic Exponential-Family Factorization Models (JMLR 2021 · similarity 0.83)
- Holographic Factorization Machines for Recommendation (AAAI 2019 · similarity 0.83)
- Polynomial Networks and Factorization Machines: New Insights and Efficient Training Algorithms (ICML 2016 · similarity 0.82)
- Personalized Time-Aware Tag Recommendation (AAAI 2018 · similarity 0.82)
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.
