arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2309.06288cs.CV
收藏

面向有限数据的自训练与多任务学习:目标检测的评估研究

Self-Training and Multi-Task Learning for Limited Data: Evaluation Study on Object Detection

IRISA(法国信息与自动化研究所);Université Bretagne Sud(南布列塔尼大学)

Hoàng-Ân Lê, Minh-Tan Pham

AI总结:

本文比较了教师数据不足时的自训练目标检测与部分标注数据下的多任务学习,实验表明弱教师模型结合未见数据训练多任务学生模型能提升性能,验证了两者在有限数据下的潜力。

摘要:

自训练允许网络从更复杂模型的预测中学习,因此通常需要训练良好的教师模型以及师生数据的混合;而多任务学习则联合优化不同目标以学习显著的相互关系,且需要每个训练样本都具有多任务标注。这些框架尽管对数据要求极高,但如果能放宽这些假设,则具有数据挖掘的潜力。本文比较了在教师训练数据不足情况下的自训练目标检测(其中学生模型由教师模型在未见样本上进行训练),以及使用部分标注数据(即每个训练样本仅有单任务标注)的多任务学习。两种场景均有其局限性,但在标注数据有限的情况下具有潜在帮助。实验结果表明,使用弱教师模型结合未见数据来训练多任务学生模型时,性能得到了提升。尽管实验设置存在局限性,我们相信实验结果展示了多任务知识蒸馏和自训练的潜力,这可能对未来的研究有所裨益。源代码位于 https://lhoangan.github.io/multas。

英文摘要:

Self-training allows a network to learn from the predictions of a more complicated model, thus often requires well-trained teacher models and mixture of teacher-student data while multi-task learning jointly optimizes different targets to learn salient interrelationship and requires multi-task annotations for each training example. These frameworks, despite being particularly data demanding have potentials for data exploitation if such assumptions can be relaxed. In this paper, we compare self-training object detection under the deficiency of teacher training data where students are trained on unseen examples by the teacher, and multi-task learning with partially annotated data, i.e. single-task annotation per training example. Both scenarios have their own limitation but potentially helpful with limited annotated data. Experimental results show the improvement of performance when using a weak teacher with unseen data for training a multi-task student. Despite the limited setup we believe the experimental results show the potential of multi-task knowledge distillation and self-training, which could be beneficial for future study. Source code is at https://lhoangan.github.io/multas.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑