arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2211.07772cs.CVcs.LGstat.ML
收藏

用于自动驾驶的鲁棒深度学习

Robust Deep Learning for Autonomous Driving

HESAM Université(HESAM大学);Conservatoire national des arts et métiers(法国国立工艺学院);Cnam(法国国立工艺学院);Télécom Paris(巴黎电信学院);INSA Rouen(鲁昂国立应用科学学院);University of Guelph(圭尔夫大学);University of Cambridge(剑桥大学);Wayve(Wayve公司);Sorbonne Université(索邦大学)

Charles Corbière

摘要:

过去十年对人工智能的研究对自动驾驶的发展产生了重大影响。然而,在高风险环境中部署此类系统时,安全性仍然是一个主要问题。本论文的目标是开发方法工具,为深度神经网络提供可靠的不确定性估计。首先,我们引入了一个新的标准来可靠地估计模型的置信度:真类概率(TCP)。我们发现,TCP提供了更好的性能故障预测比目前的不确定性措施。由于真正的类本质上是未知的,在测试时,我们建议学习TCP标准的数据与辅助模型,引入一个特定的学习计划,适应这种情况。在图像分类和语义分割数据集上验证了该方法的相关性。然后,我们扩展我们的学习信心的方法,以任务的领域自适应,它提高了自我训练方法中的伪标签的选择。最后,我们通过引入一种新的基于证据模型的不确定性度量,并在单纯形上定义,来解决联合检测误分类和分布外样本的挑战。

英文摘要:

The last decade's research in artificial intelligence had a significant impact on the advance of autonomous driving. Yet, safety remains a major concern when it comes to deploying such systems in high-risk environments. The objective of this thesis is to develop methodological tools which provide reliable uncertainty estimates for deep neural networks. First, we introduce a new criterion to reliably estimate model confidence: the true class probability (TCP). We show that TCP offers better properties for failure prediction than current uncertainty measures. Since the true class is by essence unknown at test time, we propose to learn TCP criterion from data with an auxiliary model, introducing a specific learning scheme adapted to this context. The relevance of the proposed approach is validated on image classification and semantic segmentation datasets. Then, we extend our learned confidence approach to the task of domain adaptation where it improves the selection of pseudo-labels in self-training methods. Finally, we tackle the challenge of jointly detecting misclassification and out-of-distributions samples by introducing a new uncertainty measure based on evidential models and defined on the simplex.

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。

↑