Research map / Architectures and efficiency / RNNs and normalization
Towards Faster Deep Collaborative Filtering via Hierarchical Decision Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 hierarchical rejection modules distilled from a NeuMF teacher | HDN · MovieLens-1M | NDCG@20: 0.620 -> 0.634* | Helped |
| Use hierarchical rejection modules distilled from a NeuMF teacher | HDN · Pinterest | NDCG@20: 0.193 -> 0.201* | Helped |
| Use hierarchical rejection modules distilled from a ConvNCF teacher | HDN · MovieLens-20M | HR@20: 0.866* -> 0.863 | Hurt |
| Use hierarchical rejection modules distilled from a NeuMF teacher | HDN · Yelp | NDCG@20: 0.135 -> 0.137* | Helped |
| Add distillation training rather than training HDN without distillation (ablation) | HDN · Yelp | NDCG@20: 0.129 -> 0.137* | Helped |
| Add distillation training rather than training HDN without distillation (ablation) | HDN · MovieLens-20M | NDCG@20: 0.553 -> 0.562 | Helped |
| Add item rejection during testing (ablation) | HDN · Yelp | NDCG@20: 0.131 -> 0.137 | Helped |
| Add item rejection during testing (ablation) | HDN · MovieLens-20M | NDCG@20: 0.556 -> 0.562 | Helped |
| Use teacher ranking information for rejection-layer training instead of implicit feedback alone (ablation) | HDN · Yelp | NDCG@20: 0.128 -> 0.137 | Helped |
| Use teacher ranking information for rejection-layer training instead of implicit feedback alone (ablation) | HDN · MovieLens-20M | NDCG@20: 0.549 -> 0.562 | Helped |
| Add hierarchical rejection and distillation instead of scoring all items without distillation (ablation) | HDN · Yelp | NDCG@20: 0.120 -> 0.137 | Helped |
| Add hierarchical rejection and distillation instead of scoring all items without distillation (ablation) | HDN · MovieLens-20M | NDCG@20: 0.530 -> 0.562 | Helped |
Closest papers by meaning
- A Hybrid Collaborative Filtering Model with Deep Structure for Recommender Systems (AAAI 2017 · similarity 0.90)
- Collaborative Filtering With User-Item Co-Autoregressive Models (AAAI 2018 · similarity 0.87)
- Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations (ICML 2024 · similarity 0.85)
- SHARP-Distill: A 68× Faster Recommender System with Hypergraph Neural Networks and Language Models (ICML 2025 · similarity 0.85)
- Auto Encoding Neural Process for Multi-interest Recommendation (AAAI 2025 · similarity 0.85)
- Hgformer: Hyperbolic Graph Transformer for Collaborative Filtering (ICML 2025 · similarity 0.84)
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.