Research map / Label-efficient and robust learning / Recognition and long-tail
LADA: Look-Ahead Data Acquisition via Augmentation for Deep Active Learning
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 |
|---|---|---|---|
| Integrate Var Ratio acquisition with learnable Mixup look-ahead | Active learning with ResNet-18 · FashionMNIST | test accuracy: 81.05$\pm$0.18 -> 84.47$\pm$0.89 | Helped |
| Integrate VAAL acquisition with learnable Mixup look-ahead | Active learning with ResNet-18 · SVHN | test accuracy: 75.01$\pm$0.66 -> 77.92$\pm$0.51 | Helped |
| Integrate VAAL acquisition with learnable Mixup look-ahead | Active learning with ResNet-18 · CIFAR-10 | test accuracy: 39.82$\pm$0.86 -> 44.56$\pm$1.40 | Helped |
| Integrate Max Entropy acquisition with learnable Mixup look-ahead | Active learning with ResNet-18 · FashionMNIST | test accuracy: 80.93$\pm$1.85 -> 83.67$\pm$0.29 | Helped |
| Integrate Max Entropy acquisition with learnable Mixup look-ahead | Active learning with ResNet-18 · SVHN | test accuracy: 72.57$\pm$0.76 -> 76.55$\pm$0.31 | Helped |
| Integrate Max Entropy acquisition with learnable Mixup look-ahead | Active learning with ResNet-18 · CIFAR-10 | test accuracy: 34.97$\pm$0.71 -> 37.04$\pm$1.34 | Helped |
| Add learnable Mixup policy rather than keep the policy fixed (ablation) | LADA_EntMix · FashionMNIST | test accuracy: 83.08$\pm$1.34 -> 83.67$\pm$0.29 | Helped |
| Add learnable Mixup policy rather than keep the policy fixed (ablation) | LADA_EntMix · SVHN | test accuracy: 75.73$\pm$1.48 -> 76.55$\pm$0.31 | Helped |
| Add learnable Mixup policy rather than keep the policy fixed (ablation) | LADA_EntMix · CIFAR-10 | test accuracy: 36.34$\pm$0.88 -> 37.04$\pm$1.34 | Helped |
| Add learnable Mixup policy rather than keep the policy fixed (ablation) | LADA_VaalMix · FashionMNIST | test accuracy: 82.63$\pm$0.29 -> 82.60$\pm$0.49 | No clear effect |
| Add learnable Mixup policy rather than keep the policy fixed (ablation) | LADA_VaalMix · SVHN | test accuracy: 76.83$\pm$1.05 -> 77.92$\pm$0.51 | Helped |
| Add learnable Mixup policy rather than keep the policy fixed (ablation) | LADA_VarMix · CIFAR-10 | test accuracy: 35.98$\pm$1.68 -> 36.84$\pm$0.51 | Helped |
Closest papers by meaning
- Bayesian Generative Active Deep Learning (ICML 2019 · similarity 0.86)
- Data-Efficient Augmentation for Training Neural Networks (NeurIPS 2022 · similarity 0.85)
- Towards Controlled Data Augmentations for Active Learning (ICML 2023 · similarity 0.85)
- Information Templates: A New Paradigm for Intelligent Active Feature Acquisition (NeurIPS 2026 · similarity 0.83)
- Active Learning with LLMs for Partially Observed and Cost-Aware Scenarios (NeurIPS 2024 · similarity 0.83)
- Revisiting Data Augmentation in Deep Reinforcement Learning (ICLR 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.