Zhang Chongzhen, Wang Jianrui, Yen Gary G, Zhao Chaoqiang, Sun Qiyu, Tang Yang, Qian Feng, Kurths Jürgen
Key Laboratory of Advanced Control and Optimization for Chemical Process, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China.
School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK 74075, USA.
Patterns (N Y). 2020 Jul 10;1(4):100050. doi: 10.1016/j.patter.2020.100050.
With widespread applications of artificial intelligence (AI), the capabilities of the perception, understanding, decision-making, and control for autonomous systems have improved significantly in recent years. When autonomous systems consider the performance of accuracy and transferability, several AI methods, such as adversarial learning, reinforcement learning (RL), and meta-learning, show their powerful performance. Here, we review the learning-based approaches in autonomous systems from the perspectives of accuracy and transferability. Accuracy means that a well-trained model shows good results during the testing phase, in which the testing set shares a same task or a data distribution with the training set. Transferability means that when a well-trained model is transferred to other testing domains, the accuracy is still good. Firstly, we introduce some basic concepts of transfer learning and then present some preliminaries of adversarial learning, RL, and meta-learning. Secondly, we focus on reviewing the accuracy or transferability or both of these approaches to show the advantages of adversarial learning, such as generative adversarial networks, in typical computer vision tasks in autonomous systems, including image style transfer, image super-resolution, image deblurring/dehazing/rain removal, semantic segmentation, depth estimation, pedestrian detection, and person re-identification. We furthermore review the performance of RL and meta-learning from the aspects of accuracy or transferability or both of them in autonomous systems, involving pedestrian tracking, robot navigation, and robotic manipulation. Finally, we discuss several challenges and future topics for the use of adversarial learning, RL, and meta-learning in autonomous systems.
随着人工智能（AI）的广泛应用，近年来自主系统在感知、理解、决策和控制方面的能力有了显著提升。当自主系统考虑准确性和可迁移性时，一些人工智能方法，如对抗学习、强化学习（RL）和元学习，展现出强大的性能。在此，我们从准确性和可迁移性的角度回顾自主系统中基于学习的方法。准确性意味着一个训练良好的模型在测试阶段表现出良好的结果，其中测试集与训练集共享相同的任务或数据分布。可迁移性意味着当一个训练良好的模型被转移到其他测试领域时，准确性仍然良好。首先，我们介绍一些迁移学习的基本概念，然后阐述对抗学习、强化学习和元学习的一些预备知识。其次，我们重点回顾这些方法在准确性或可迁移性或两者方面的表现，以展示对抗学习（如生成对抗网络）在自主系统典型计算机视觉任务中的优势，包括图像风格迁移、图像超分辨率、图像去模糊/去雾/除雨、语义分割、深度估计、行人检测和行人重识别。我们还从准确性或可迁移性或两者方面回顾了强化学习和元学习在自主系统中的性能，涉及行人跟踪、机器人导航和机器人操作。最后，我们讨论了在自主系统中使用对抗学习、强化学习和元学习的几个挑战和未来研究课题。
