Research map / Graph learning / Graph algorithms and clustering
Learning to Multitask
Yu Zhang, Ying Wei, Qiang Yang
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 |
|---|---|---|---|
| learn the task covariance from data rather than imposing an explicit task-covariance assumption | L2MT · MIT-Indoor-Scene, Caltech256, 20newsgroup, and RCV1 | relative test error | Improved |
| fine-tune the VGG-19 FC layers using the multitask objective (ablation) | L2MT · Caltech256 and MIT-Indoor-Scene | average test error: reduced by about 5% | Improved |
| replace the proposed estimation and link functions with the alternative formulations (ablation) | L2MT · Benchmark datasets | test error: about 2% relative increase on the test error | Hurt |
| increase LGNN depth from 1 layer to 2 or 3 layers (ablation) | L2MT · 20newsgroup | relative test error | Improved |
| increase LGNN depth beyond 3 layers (ablation) | L2MT · 20newsgroup | relative test error | Hurt |
| increase the LGNN latent dimension to a larger value (ablation) | L2MT · 20newsgroup | relative test error | Hurt |
| increase the number of historical multitask problems used for training (ablation) | L2MT · 20newsgroup | test error | Improved |
Closest papers by meaning
- The Benefit of Multitask Representation Learning (JMLR 2016 · similarity 0.86)
- Asymmetric Multi-task Learning Based on Task Relatedness and Loss (ICML 2016 · similarity 0.85)
- Multiplicative Multitask Feature Learning (JMLR 2016 · similarity 0.85)
- Learning Multi-Task Communication with Message Passing for Sequence Learning (AAAI 2019 · similarity 0.85)
- NTKMTL: Mitigating Task Imbalance in Multi-Task Learning from Neural Tangent Kernel Perspective (NeurIPS 2025 · similarity 0.84)
- Provable Multi-Task Representation Learning by Two-Layer ReLU Neural Networks (ICML 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.
