Research map / Learning theory and algorithms / Prediction and decision losses
GLISTER: Generalization based Data Subset Selection for Efficient and Robust 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 |
|---|---|---|---|
| Select a 10% training subset with GLISTER instead of full training | ResNet-18 · CIFAR-10 | speedup: 6x | Helped |
| Select a 10% training subset with GLISTER instead of full training | ResNet-18 · CIFAR-10 | accuracy: losing 3% | Hurt |
| Select a 30% training subset with GLISTER instead of full training | ResNet-18 · CIFAR-10 | speedup: 2.5x | Helped |
| Select a 30% training subset with GLISTER instead of full training | ResNet-18 · CIFAR-10 | accuracy: 1.2% | Hurt |
| Select a 50% training subset with GLISTER instead of full training | ResNet-18 · CIFAR-10 | speedup: 1.5x | Helped |
| Select a 50% training subset with GLISTER instead of full training | ResNet-18 · CIFAR-10 | accuracy: 0.2% | Hurt |
| Select a 10% training subset with GLISTER instead of full training | LeNet · MNIST | speedup: 3x | Helped |
| Select a 10% training subset with GLISTER instead of full training | LeNet · MNIST | accuracy: 0.2% | Hurt |
| Use GLISTER data selection | Two-layer neural network · DNA | speedup: 6.75x | Helped |
| Use GLISTER data selection | Two-layer neural network · SVMGuide | speedup: 4.9x | Helped |
| Use GLISTER data selection | Two-layer neural network · SatImage | speedup: 4.6x | Helped |
| Use GLISTER data selection | Two-layer neural network · USPS and Letter | speedup: around 1.5x | Helped |
Closest papers by meaning
- GIO: Gradient Information Optimization for Training Dataset Selection (ICLR 2024 · similarity 0.85)
- Most Influential Subset Selection: Challenges, Promises, and Beyond (NeurIPS 2024 · similarity 0.85)
- Improving Generalization via Meta-Learning on Hard Samples (CVPR 2024 · similarity 0.84)
- SAND: One-Shot Feature Selection with Additive Noise Distortion (ICML 2025 · similarity 0.84)
- Data-Efficient Learning via Clustering-Based Sensitivity Sampling: Foundation Models and Beyond (ICML 2024 · similarity 0.84)
- PEAKS: Selecting Key Training Examples Incrementally via Prediction Error Anchored by Kernel Similarity (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.