SIMS: Scale-Invariant Merit-Function-Based Scalarization for Multi-Task Learning
SIMS adopts a transformation-induced merit function to convert the MOO problem of MTL to a single objective that renders optimization invariant to the magnitudes of losses, and proves that the requirement for scale invariance uniquely determines this transformation to be logarithmic.
Ze-Bin Chen, Fei Xing, Yang Chen et al.
· Proceedings of the 32nd ACM... · 0 citations