A new first-order algorithm for multi-task learning with shared linear representation achieves near-optimal error rates in constant iterations, improving existing methods by a factor of k.
National Science Review , volume=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Introduces a regularized estimator achieving optimal MSE rates under a new relative balancedness condition while providing safety guarantees that match independent learning when tasks are unrelated.
citing papers explorer
-
Near-optimal and Efficient First-Order Algorithm for Multi-Task Learning with Shared Linear Representation
A new first-order algorithm for multi-task learning with shared linear representation achieves near-optimal error rates in constant iterations, improving existing methods by a factor of k.
-
Multi-task Linear Regression without Eigenvalue Lower Bounds: Adaptivity, Robustness and Safety
Introduces a regularized estimator achieving optimal MSE rates under a new relative balancedness condition while providing safety guarantees that match independent learning when tasks are unrelated.