Research map / Learning theory and algorithms / Kernels and regression theory
Generalization in multitask deep neural classifiers: a statistical physics approach
Anthony Ndirango, Tyler Lee
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 |
|---|---|---|---|
| Jointly train with a related, clean auxiliary task | Multitask softmax classifier · Main task A | generalization loss | Improved |
| Jointly train with an unrelated, high-SNR auxiliary task | Multitask softmax classifier · Main task A | generalization loss | Hurt |
| Jointly train with an unrelated, noisy auxiliary task | Multitask softmax classifier · Main task A | generalization loss | No clear effect |
| Increase auxiliary-task data while holding main-task data fixed | Multitask student network · Main task A with 100 main-task data points | multitask benefit | Mixed |
| Increase task relatedness | Multitask student network · Main task A | multitask benefit | Improved |
| Increase auxiliary-task SNR | Multitask student network · Main task A | multitask benefit | Mixed |
| Provide abundant labeled training data | Multitask softmax classifier · Main task A | multitask benefit | No clear effect |
| Use deeper student networks with ReLU nonlinearities | Multitask student network · Multitask generalization | multitask benefit | Hurt |
Closest papers by meaning
- Provable Multi-Task Representation Learning by Two-Layer ReLU Neural Networks (ICML 2024 · similarity 0.88)
- Algorithm Development in Neural Networks: Insights from the Streaming Parity Task (ICML 2025 · similarity 0.88)
- Topological Generalization Bounds for Discrete-Time Stochastic Optimization Algorithms (NeurIPS 2024 · similarity 0.87)
- Analyzing Neural Scaling Laws in Two-Layer Networks with Power-Law Data Spectra (ICLR 2025 · similarity 0.87)
- Scaling can lead to compositional generalization (NeurIPS 2025 · similarity 0.87)
- Grokking at the Edge of Numerical Stability (ICLR 2025 · 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.
