When Does Multi-Task Learning Fail? Quantifying Data Imbalance and Task Independence in Metal Alloy Property Prediction
By: Sungwoo Kang
Multi-task learning (MTL) assumes related material properties share underlying physics that can be leveraged for better predictions. We test this by simultaneously predicting electrical resistivity, Vickers hardness, and amorphous-forming ability using 54,028 alloy samples. We compare single-task models against standard and structured MTL. Results reveal a striking dichotomy: MTL significantly degrades regression performance (resistivity $R^2$: 0.897 $\to$ 0.844; hardness $R^2$: 0.832 $\to$ 0.694, $p < 0.01$) but improves classification (amorphous F1: 0.703 $\to$ 0.744, $p < 0.05$; recall +17%). Analysis shows near-zero inter-task weights, indicating property independence. Regression failure is attributed to negative transfer caused by severe data imbalance (52k vs. 800 samples). We recommend independent models for precise regression, while reserving MTL for classification tasks where recall is critical.
Similar Papers
Tensorized Multi-Task Learning for Personalized Modeling of Heterogeneous Individuals with High-Dimensional Data
Machine Learning (CS)
Helps computers learn about different groups of people.
Multi-task Learning for Heterogeneous Data via Integrating Shared and Task-Specific Encodings
Machine Learning (Stat)
Helps doctors predict cancer growth better.
FairMT: Fairness for Heterogeneous Multi-Task Learning
Machine Learning (CS)
Makes AI fair for different jobs and missing info.