Research map / Architectures and efficiency / Neural architecture search
Optimizing Network Simulation: Enhancing Performance Prediction Accuracy via Neural Architecture Search
ShaoChen He, Zirui Zhuang, Haifeng Sun, Xiaoyuan Fu, Qi Qi, Lei Zhang, Jianxin Liao, Jingyu Wang
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 |
|---|---|---|---|
| add Wasserstein loss to MSE training (ablation) | ANAS · DeepQueueNet switch dataset (test) | p99Wn (x1e-3): 18.1 -> 17.0 | Improved |
| add Wasserstein loss to MSE training (ablation) | ANAS · DeepQueueNet switch dataset (test) | MSE (x1e-5): 5.32 -> 5.14 | Improved |
| add similarity-guided evaluation and dynamic warm-up (ablation) | ANAS · DeepQueueNet switch dataset (test) | MSE (x1e-5): 5.78 -> 5.14 | Improved |
| use hybrid attention-recurrent encoder-decoder search space instead of cell-based space (ablation) | ANAS · DeepQueueNet switch dataset (test) | MSE (x1e-5): 5.94 -> 5.14 | Improved |
| ANAS search strategy vs double-budget random search in same space and loss (ablation) | ANAS · DeepQueueNet switch dataset (test) | MSE (x1e-5): 5.64 -> 5.14 (32h -> 16h) | Improved |
| replace manual DeepQueueNet with ANAS-discovered architecture | DeepQueueNet · DeepQueueNet switch dataset (test) | MSE (x1e-5): 6.93 -> 5.14 | Improved |
| use general-purpose Transformer instead of ANAS architecture | Transformer · DeepQueueNet switch dataset (test) | p99Wn (x1e-3): 17.0 -> 22.1 | Hurt |
| use ENAS (standard weight-sharing NAS) instead of ANAS under distribution shift | ENAS · Roblox trace end-to-end | p99Delay Wn: 0.087 -> 0.857 | Hurt |
| lower inter-architecture similarity between training and evaluated architecture (ablation) | weight-sharing supernet evaluation · Supernet evaluation L1 loss | mean L1 loss (x1e-3): 9.6 -> 13.7 | Hurt |
| use ANAS in High Load scenario vs DeepQueueNet | ANAS · High Load (HL) end-to-end | avgDelay Wn: 0.034 -> 0.026 | Improved |
Closest papers by meaning
- IS-DARTS: Stabilizing DARTS through Precise Measurement on Candidate Importance (AAAI 2024 · similarity 0.86)
- An Analysis of Super-Net Heuristics in Weight-Sharing NAS (TPAMI 2021 · similarity 0.86)
- HOTNAS: Hierarchical Optimal Transport for Neural Architecture Search (CVPR 2023 · similarity 0.85)
- AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search (CVPR 2024 · similarity 0.85)
- Surprisingly Strong Performance Prediction with Neural Graph Features (ICML 2024 · similarity 0.84)
- Subnet-Aware Dynamic Supernet Training for Neural Architecture Search (CVPR 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.
