Research map / Architectures and efficiency / Neural architecture search
Progressive Feature Interaction Search for Deep Sparse Network
Chen Gao, Yinfeng Li, quanming yao, Depeng Jin, Yong Li
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 |
|---|---|---|---|
| replace progressive search with one-step search (ablation) | PROFIT · Criteo | AUC: 0.8095 -> 0.8077 | Hurt |
| replace progressive search with one-step search (ablation) | PROFIT · Avazu | AUC: 0.7883 -> 0.7851 | Hurt |
| replace progressive search with one-step search (ablation) | PROFIT · ML1M | AUC: 0.8077 -> 0.8053 | Hurt |
| replace progressive search with random search with parameter sharing (ablation) | PROFIT · Criteo | AUC: 0.8095 -> 0.7718 | Hurt |
| replace progressive search with random search with parameter sharing (ablation) | PROFIT · Avazu | AUC: 0.7883 -> 0.7499 | Hurt |
| replace progressive search with random search with parameter sharing (ablation) | PROFIT · ML1M | AUC: 0.8077 -> 0.7692 | Hurt |
| search in full space instead of distilled space (ablation) | PROFIT · Criteo | AUC: 0.8095 -> 0.7975 | Hurt |
| search in full space instead of distilled space (ablation) | PROFIT · Avazu | AUC: 0.7883 -> 0.7807 | Hurt |
| search in full space instead of distilled space (ablation) | PROFIT · ML1M | AUC: 0.8077 -> 0.7957 | Hurt |
| search in full space instead of distilled space (ablation) | PROFIT · Criteo | Logloss: 0.4427 -> 0.4533 | Hurt |
| search in full space instead of distilled space (ablation) | PROFIT · Avazu | Logloss: 0.3735 -> 0.3802 | Hurt |
| search in full space instead of distilled space (ablation) | PROFIT · ML1M | Logloss: 0.5281 -> 0.5353 | Hurt |
Closest papers by meaning
- Towards Hybrid-grained Feature Interaction Selection for Deep Sparse Network (NeurIPS 2023 · similarity 0.88)
- Sparse Interaction Additive Networks via Feature Interaction Detection and Sparse Selection (NeurIPS 2022 · similarity 0.87)
- Random Sparse Lifts: Construction, Analysis and Convergence of finite sparse networks (ICLR 2024 · similarity 0.84)
- Sparse maximal update parameterization: A holistic approach to sparse training dynamics (NeurIPS 2024 · similarity 0.83)
- Sparse Cocktail: Every Sparse Pattern Every Sparse Ratio All At Once (ICML 2024 · similarity 0.83)
- Matching the Statistical Query Lower Bound for $k$-Sparse Parity Problems with Sign Stochastic Gradient Descent (NeurIPS 2024 · similarity 0.83)
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.
