Multi-Task Learning with User Preferences: Gradient Descent with Controlled Ascent in Pareto Optimization.

Debabrata Mahapatra, Vaibhav Rajan

Browse the full ICML paper archive.