Multi-Task Learning for Robot Perception with Imbalanced Data

Home Page
About
Submit A Journal
Submit A Conference
Submit Paper/Book
- Submit a Preprint
- Submit a Book
Contact

Ordu Üniversitesi Bilim ve Teknoloji Dergisi
Cilt: 15 Sayı: 2
Multi-Task Learning for Robot Perception with Imbalanced Data

Multi-Task Learning for Robot Perception with Imbalanced Data

Pages : 151-164

View : 81 | Download : 153

Publication Date : 2025-12-27

Article Type : Research Paper

Abstract :Multi-task problem solving has been shown to improve the accuracy of the individual tasks, which is an important feature for robots, as they have a limited resource. However, when the number of labels for each task is not equal, namely imbalanced data exists, a problem may arise due to insufficient number of samples, and labeling is not very easy for mobile robots in every environment. We propose a method that can learn tasks even in the absence of ground truth labels for some of the tasks. We also provide a detailed analysis of the proposed method. An interesting finding is related to the interaction of the tasks. We show a methodology to find out which tasks can improve the performance of other tasks. We investigate this by training the teacher network with the task outputs such as depth as inputs. We further provide empirical evidence when trained with a small amount of data. We use semantic segmentation and depth estimation tasks on different datasets, NYUDv2 and Cityscapes.
Keywords : otonom robotlar, bölütleme, derinlik tahmini, çok görevli öğrenmede dengesiz veriler

ORIGINAL ARTICLE URL

* There may have been changes in the journal, article,conference, book, preprint etc. informations. Therefore, it would be appropriate to follow the information on the official page of the source. The information here is shared for informational purposes. IAD is not responsible for incorrect or missing information.