Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Adaptive and Robust Multi-Task Learning

About

We study the multi-task learning problem that aims to simultaneously analyze multiple datasets collected from different sources and learn one model for each of them. We propose a family of adaptive methods that automatically utilize possible similarities among those tasks while carefully handling their differences. We derive sharp statistical guarantees for the methods and prove their robustness against outlier tasks. Numerical experiments on synthetic and real datasets demonstrate the efficacy of our new methods.

Yaqi Duan, Kaizheng Wang• 2022

Related benchmarks

TaskDatasetResultRank
Linear regressionSynthetic Linear Regression (n=50, d=50, K=40, ε=0.2) (train/test)
Average Local Error0.389
204
Linear regressionSynthetic Multi-task Linear Regression (n=d=50, K=40, epsilon=0.2)
Global Error (L2-norm)0.39
192
Multi-task Linear Regression (Local Parameter Estimation)Synthetic Multi-task Linear Regression (n=d=50, ε=0.2) (test)
Local L2 Error1.173
102
Local predictionHAR (20% train)
Local Prediction Error3.1
102
Local prediction errorHAR 50% (train)
Error Rate2
102
Local predictionHAR 60% (train)
Local Prediction Error1.9
102
Linear regressionSynthetic Multi-task Dataset d=50, K=20, epsilon=0.2 (test)
Local Error1.091
102
Multi-task Logistic RegressionSynthetic Multi-task Logistic Regression (n=d=50, epsilon=0.2)
Local Error2.695
102
Logistic RegressionLogistic regression d=50, K=20, epsilon=0.2 varying per-task sample size n (test)
Local Error2.726
102
Linear regressionLinear Regression (synthetic)
Local Error0.984
101
Showing 10 of 14 rows

Other info

Follow for update