Problem
Research questions and friction points this paper is trying to address.
unsupervised domain adaptation
multitask image analysis
extreme label shift
class imbalance
domain shift
Unsupervised Domain Adaptation for Multitask Image Analysis in Realistic Context with Extreme Label Shift; Application to the CTAO first Large Sized Telescope
🤖 AI Summary
This work addresses the significant performance degradation observed when combining unsupervised domain adaptation with multi-task learning under extreme label shift. To tackle this challenge, the authors propose a unified framework that jointly optimizes domain adaptation and multi-task balancing in realistic high-energy physics imaging scenarios, such as those encountered in Cherenkov Telescope Array Observatory (CTAO) observations. The approach incorporates auxiliary adaptation tasks, a task-weight scheduling mechanism, and an extended importance weighting scheme to correct distributional bias in the target domain. Experiments on image data from the first large-sized telescope of the CTAO demonstrate that the proposed framework effectively enhances target-domain performance. The implementation code and experimental results have been made publicly available on Zenodo.
📝 Abstract
Unsupervised domain adaptation is a widespread set of methods that leverages the knowledge of a labeled source domain to train a model to perform well on a related unlabeled target domain. They generally introduce an auxiliary adaptation-related task that can be integrated into the multitask paradigm, which aims to merge multiple single-task models into a unified architecture. In this paper, we propose to associate domain adaptation and multitask balancing in the realistic context of an extreme class imbalance. Therefore, we propose a combined framework to cover and validate these approaches, and evaluate its performance in the physics-based context of the Cherenkov Telescope Array Observatory (CTAO). Along with a comparative study of some relevant adaptation techniques, we highlight the impact of extreme label shift and extend the investigations on importance weighting to rectify it. The complete code and results are published and available as open-source resources on Zenodo.
Innovation
Methods, ideas, or system contributions that make the work stand out.
unsupervised domain adaptation
multitask learning
extreme label shift
importance weighting
Cherenkov Telescope Array Observatory
🔎 Similar Papers
No similar papers found.
