Learning-Deep-Learning

Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics

August 2019

tl;dr: Self-paced learning based on homoscedastic uncertainty.

Overall impression

The paper spent much math details deriving the formulation of mutitask loss function based on the idea of maximizing the Gaussian likelihood with himoscedastic uncertainty. However once implemented, the formulation is extremely straightforward and easy to implement.

Methods Learning Progress Signal hyperparameters
Uncertainty Weighting Homoscedastic Uncertainty No hyperparameters
GradNorm Training loss ratio 1 exponential weighting factor
Dynamic Task Prioritization KPI 1 focal loss scaling factor

Key ideas

Technical details

Notes