HydaLearn: Highly Dynamic Task Weighting for Multi-task Learning with Auxiliary Tasks

08/26/2020
by   Sam Verboven, et al.
0

Multi-task learning (MTL) can improve performance on a task by sharing representations with one or more related auxiliary-tasks. Usually, MTL-networks are trained on a composite loss function formed by a constant weighted combination of the separate task losses. In practice, constant loss weights lead to poor results for two reasons: (i) the relevance of the auxiliary tasks can gradually drift throughout the learning process; (ii) for mini-batch based optimisation, the optimal task weights vary significantly from one update to the next depending on mini-batch sample composition. We introduce HydaLearn, an intelligent weighting algorithm that connects main-task gain to the individual task gradients, in order to inform dynamic loss weighting at the mini-batch level, addressing i and ii. Using HydaLearn, we report performance increases on synthetic data, as well as on two supervised learning domains.

READ FULL TEXT
research
06/07/2023

Sample-Level Weighting for Multi-Task Learning with Auxiliary Tasks

Multi-task learning (MTL) can improve the generalization performance of ...
research
09/03/2020

Multi-Loss Weighting with Coefficient of Variations

Many interesting tasks in machine learning and computer vision are learn...
research
05/19/2017

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

Numerous deep learning applications benefit from multi-task learning wit...
research
12/05/2018

Adapting Auxiliary Losses Using Gradient Similarity

One approach to deal with the statistical inefficiency of neural network...
research
02/07/2022

Auto-Lambda: Disentangling Dynamic Task Relationships

Understanding the structure of multiple related tasks allows for multi-t...
research
11/20/2021

A Closer Look at Loss Weighting in Multi-Task Learning

Multi-Task Learning (MTL) has achieved great success in various fields, ...
research
09/07/2019

Road Mapping In LiDAR Images Using A Joint-Task Dense Dilated Convolutions Merging Network

It is important, but challenging, for the forest industry to accurately ...

Please sign up or login with your details

Forgot password? Click here to reset