Research map / Label-efficient and robust learning / Recognition and long-tail
Learning New Tricks From Old Dogs: Multi-Source Transfer Learning From Pre-Trained Networks
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 |
|---|---|---|---|
| use multiple source networks instead of the best single source | MCW · CIFAR-100 apple vs fish, 5-shot | accuracy: 69.0 -> 78.1 | Helped |
| MCW instead of SVM on multi-source features | Multi-source features · CIFAR-100 apple vs fish, 1-shot | accuracy: 64.7 -> 69.0 | Helped |
| MCW instead of SVM on multi-source features | Multi-source features · CIFAR-100 apple vs fish, 20-shot | accuracy: 81.5 -> 81.7 | No clear effect |
| MCW instead of SVM on multi-source features | Multi-source features · Stanford Dogs 5-way, 5-shot | accuracy: 38.9 -> 41.6 | Helped |
| MCW instead of SVM on multi-source features | Multi-source features · Tiny ImageNet 5-way, 5-shot | accuracy: 42.5 -> 47.4 | Helped |
| use multiple sources instead of best single source | MCW · Tiny ImageNet 5-way, 5-shot | accuracy: 33.9 -> 47.4 | Helped |
| use multiple sources instead of best single source | MCW · Stanford Dogs 5-way, 5-shot | accuracy: 38.2 -> 41.6 | Helped |
| add least-favoured source task (camel vs can) (ablation) | MCW with 9 sources · CIFAR-100 apple vs fish, 5-shot | accuracy: 76.8 -> 78.1 | No clear effect |
| add most-favoured source task (dolphin vs elephant) (ablation) | MCW with 9 sources · CIFAR-100 apple vs fish, 5-shot | accuracy: 73.0 -> 78.1 | Helped |
| MCW instead of SVM on best single source | Single-source features · CIFAR-100 apple vs fish, 10-shot | accuracy: 70.4 -> 67.0 | Hurt |
Closest papers by meaning
- H-ensemble: An Information Theoretic Approach to Reliable Few-Shot Multi-Source-Free Transfer (AAAI 2024 · similarity 0.88)
- A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning (NeurIPS 2025 · similarity 0.88)
- A Mathematical Framework for Quantifying Transferability in Multi-source Transfer Learning (NeurIPS 2021 · similarity 0.87)
- Adapted Deep Embeddings: A Synthesis of Methods for k-Shot Inductive Transfer Learning (NeurIPS 2018 · similarity 0.86)
- Transferring Knowledge From Large Foundation Models to Small Downstream Models (ICML 2024 · similarity 0.85)
- Diffusion-based Neural Network Weights Generation (ICLR 2025 · similarity 0.85)
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.