Research map / Architectures and efficiency / Neural architecture search
IDARTS: Interactive Differentiable Architecture Search
Song Xue, Runqi Wang, Baochang Zhang, Tian Wang, Guodong Guo, David Doermann
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 backtracking of architecture parameters (IDARTS) on top of PC-DARTS-style search | PC-DARTS · CIFAR10 test | accuracy (%): 97.43 -> 97.68 | Improved |
| add backtracking of architecture parameters (IDARTS) on top of PC-DARTS-style search | PC-DARTS · ImageNet val (searched on ImageNet) | top-1 accuracy (%): 75.8 -> 76.52 | Improved |
| add backtracking of architecture parameters (IDARTS) | PC-DARTS · ImageNet val (searched on CIFAR10) | top-1 accuracy (%): 74.9 -> 76.14 | Improved |
| IDARTS versus CDARTS (cyclic feedback search) | CDARTS · CIFAR10 test | accuracy (%): 97.52 -> 97.68 | Improved |
| IDARTS versus CDARTS search cost | CDARTS · CIFAR10 test | search cost (GPU days): 0.3 -> 0.1 | Improved |
| search directly on ImageNet instead of transferring a CIFAR10-searched cell | IDARTS · ImageNet val | top-1 accuracy (%): 76.14 -> 76.52 | Improved |
| search directly on ImageNet instead of CIFAR10 (search cost) | IDARTS · ImageNet val | search cost (GPU days): 0.1 -> 3.8 | Hurt |
| IDARTS versus ProxylessNAS | ProxylessNAS · CIFAR10 test | accuracy (%): 97.92 -> 97.68 | Hurt |
| IDARTS versus manual DenseNet-BC | DenseNet-BC · CIFAR10 test | accuracy (%): 96.54 -> 97.68 | Improved |
| IDARTS versus manual ShuffleNet 2x (v2) | ShuffleNet 2x (v2) · ImageNet val | top-1 accuracy (%): 74.9 -> 76.52 | Improved |
| add backtracking so alpha escapes local minima during search (ablation) | PC-DARTS · CIFAR10 search | final search validation loss | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FairDARTS — Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture Search (ECCV 2020) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients (ICML 2021 · similarity 0.91)
- Cyclic Differentiable Architecture Search (TPAMI 2022 · similarity 0.90)
- DARTS: Differentiable Architecture Search (ICLR 2019 · similarity 0.90)
- PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search (ICLR 2020 · similarity 0.90)
- EC-DARTS: Inducing Equalized and Consistent Optimization Into DARTS (ICCV 2021 · similarity 0.90)
- EG-NAS: Neural Architecture Search with Fast Evolutionary Exploration (AAAI 2024 · similarity 0.87)
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.
