Resource Type

Journal Article 503

Conference Information 44

Conference Videos 24

Conference Topics 2

Year

2024 2

2023 78

2022 74

2021 70

2020 68

2019 76

2018 32

2017 32

2016 16

2015 10

2014 6

2013 5

2012 7

2011 4

2010 6

2009 5

2008 8

2007 12

2006 9

2005 8

open ︾

Keywords

Machine learning 42

Deep learning 34

Artificial intelligence 15

Reinforcement learning 14

Autonomous driving 5

Active learning 4

Process intensification 4

Bayesian optimization 3

Big data 3

Adaptive dynamic programming 2

Additive manufacturing 2

Attention 2

Autonomous learning 2

Autonomous vehicle 2

Chemical engineering 2

Data-driven 2

Feature learning 2

Feature selection 2

Materials design 2

open ︾

Search scope:

排序: Display mode:

Pre-training with asynchronous supervised learning for reinforcement learning based autonomous driving Research Articles

Yunpeng Wang, Kunxian Zheng, Daxin Tian, Xuting Duan, Jianshan Zhou,ypwang@buaa.edu.cn,zhengkunxian@buaa.edu.cn,dtian@buaa.edu.cn,duanxuting@buaa.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 5,   Pages 615-766 doi: 10.1631/FITEE.1900637

Abstract: Rule-based autonomous driving systems may suffer from increased complexity with large-scale inter-coupled rules, so many researchers are exploring learning-based approaches. (RL) has been applied in designing autonomous driving systems because of its outstanding performance on a wide variety of sequential control problems. However, poor initial performance is a major challenge to the practical implementation of an RL-based autonomous driving system. RL training requires extensive training data before the model achieves reasonable performance, making an RL-based model inapplicable in a real-world setting, particularly when data are expensive. We propose an asynchronous (ASL) method for the RL-based end-to-end autonomous driving model to address the problem of poor initial performance before training this RL-based model in real-world settings. Specifically, prior knowledge is introduced in the ASL pre-training stage by asynchronously executing multiple processes in parallel, on multiple driving demonstration data sets. After pre-training, the model is deployed on a real vehicle to be further trained by RL to adapt to the real environment and continuously break the performance limit. The presented pre-training method is evaluated on the race car simulator, TORCS (The Open Racing Car Simulator), to verify that it can be sufficiently reliable in improving the initial performance and convergence speed of an end-to-end autonomous driving model in the RL training stage. In addition, a real-vehicle verification system is built to verify the feasibility of the proposed pre-training method in a real-vehicle deployment. Simulations results show that using some demonstrations during a supervised pre-training stage allows significant improvements in initial performance and convergence speed in the RL training stage.

Keywords: 自主驾驶;自动驾驶车辆;强化学习;监督学习    

Toward Human-in-the-loop AI: Enhancing Deep Reinforcement Learning Via Real-time Human Guidance for Autonomous Driving Article

Jingda Wu, Zhiyu Huang, Zhongxu Hu, Chen Lv

Engineering 2023, Volume 21, Issue 2,   Pages 75-91 doi: 10.1016/j.eng.2022.05.017

Abstract:

Due to its limited intelligence and abilities, machine learning is currently unable to handle various situations thus cannot completely replace humans in real-world applications. Because humans exhibit robustness and adaptability in complex scenarios, it is crucial to introduce humans into the training loop of artificial intelligence (AI), leveraging human intelligence to further advance machine learning algorithms. In this study, a real-time human-guidance-based (Hug)-deep reinforcement learning (DRL) method is developed for policy training in an end-to-end autonomous driving case. With our newly designed mechanism for control transfer between humans and automation, humans are able to intervene and correct the agent's unreasonable actions in real time when necessary during the model training process. Based on this human-in-the-loop guidance mechanism, an improved actor-critic architecture with modified policy and value networks is developed. The fast convergence of the proposed Hug-DRL allows real-time human guidance actions to be fused into the agent's training loop, further improving the efficiency and performance of DRL. The developed method is validated by human-in-the-loop experiments with 40 subjects and compared with other state-of-the-art learning approaches. The results suggest that the proposed method can effectively enhance the training efficiency and performance of the DRL algorithm under human guidance without imposing specific requirements on participants' expertise or experience.

Keywords: Human-in-the-loop AI     Deep reinforcement learning     Human guidance     Autonomous driving    

Stochastic pedestrian avoidance for autonomous vehicles using hybrid reinforcement learning Research Article

Huiqian LI, Jin HUANG, Zhong CAO, Diange YANG, Zhihua ZHONG,lihq20@mails.tsinghua.edu.cn,huangjin@tsinghua.edu.cn,caoc15@mails.tsinghua.edu.cn,ydg@tsinghua.edu.cn

Frontiers of Information Technology & Electronic Engineering 2023, Volume 24, Issue 1,   Pages 131-140 doi: 10.1631/FITEE.2200128

Abstract: Ensuring the safety of s is essential and challenging when are involved. Classical avoidance strategies cannot handle uncertainty, and learning-based methods lack performance guarantees. In this paper we propose a (HRL) approach for to safely interact with s behaving uncertainly. The method integrates the rule-based strategy and reinforcement learning strategy. The confidence of both strategies is evaluated using the data recorded in the training process. Then we design an activation function to select the final policy with higher confidence. In this way, we can guarantee that the final policy performance is not worse than that of the rule-based policy. To demonstrate the effectiveness of the proposed method, we validate it in simulation using an accelerated testing technique to generate stochastic s. The results indicate that it increases the success rate for avoidance to 98.8%, compared with 94.4% of the baseline method.

Keywords: Pedestrian     Hybrid reinforcement learning     Autonomous vehicles     Decision-making    

Towards the Unified Principles for Level 5 Autonomous Vehicles Article

Jianqiang Wang, Heye Huang, Keqiang Li, Jun Li

Engineering 2021, Volume 7, Issue 9,   Pages 1313-1325 doi: 10.1016/j.eng.2020.10.018

Abstract:

The rapid advance of autonomous vehicles (AVs) has motivated new perspectives and potential challenges for existing modes of transportation. Currently, driving assistance systems of Level 3 and below have been widely produced, and several applications of Level 4 systems to specific situations have also been gradually developed. By improving the automation level and vehicle intelligence, these systems
can be further advanced towards fully autonomous driving. However, general development concepts for Level 5 AVs remain unclear, and the existing methods employed in the development processes of Levels 0–4 have been mainly based on task-driven function development related to specific scenarios. Therefore, it is difficult to identify the problems encountered by high-level AVs. The essential logical
and physical mechanisms of vehicles have hindered further progression towards Level 5 systems. By exploring the physical mechanisms behind high-level autonomous driving systems and analyzing the essence of driving, we put forward a coordinated and balanced framework based on the brain–cerebellum–organ concept through reasoning and deduction. Based on a mixed mode relying on the crow inference and parrot imitation approach, we explore the research paradigm of autonomous learning and prior knowledge to realize the characteristics of self-learning, self-adaptation, and self-transcendence for AVs. From a systematic, unified, and balanced point of view and based on least action principles and unified safety field concepts, we aim to provide a novel research concept and develop an effective approach for the research and development of high-level AVs, specifically at Level 5.

Keywords: Autonomous vehicle     Principle of least action     Driving safety field     Autonomous learning     Basic paradigm    

Embedding expert demonstrations into clustering buffer for effective deep reinforcement learning Research Article

Shihmin WANG, Binqi ZHAO, Zhengfeng ZHANG, Junping ZHANG, Jian PU

Frontiers of Information Technology & Electronic Engineering 2023, Volume 24, Issue 11,   Pages 1541-1556 doi: 10.1631/FITEE.2300084

Abstract: As one of the most fundamental topics in (RL), is essential to the deployment of deep RL algorithms. Unlike most existing exploration methods that sample an action from different types of posterior distributions, we focus on the policy and propose an efficient selective sampling approach to improve by modeling the internal hierarchy of the environment. Specifically, we first employ in the policy to generate an action candidate set. Then we introduce a clustering buffer for modeling the internal hierarchy, which consists of on-policy data, off-policy data, and expert data to evaluate actions from the clusters in the action candidate set in the exploration stage. In this way, our approach is able to take advantage of the supervision information in the expert demonstration data. Experiments on six different continuous locomotion environments demonstrate superior performance and faster convergence of selective sampling. In particular, on the LGSVL task, our method can reduce the number of convergence steps by 46.7% and the convergence time by 28.5%. Furthermore, our code is open-source for reproducibility. The code is available at https://github.com/Shihwin/SelectiveSampling.

Keywords: Reinforcement learning     Sample efficiency     Sampling process     Clustering methods     Autonomous driving    

A Probabilistic Architecture of Long-Term Vehicle Trajectory Prediction for Autonomous Driving Article

Jinxin Liu, Yugong Luo, Zhihua Zhong, Keqiang Li, Heye Huang, Hui Xiong

Engineering 2022, Volume 19, Issue 12,   Pages 228-239 doi: 10.1016/j.eng.2021.12.020

Abstract:

In mixed and dynamic traffic environments, accurate long-term trajectory forecasting of surrounding vehicles is one of the indispensable preconditions for autonomous vehicles (AVs) to accomplish reasonable behavioral decisions and guarantee driving safety. In this paper, we propose an integrated probabilistic architecture for long-term vehicle trajectory prediction, which consists of a driving inference model (DIM) and a trajectory prediction model (TPM). The DIM is designed and employed to accurately infer the potential driving intention based on a dynamic Bayesian network. The proposed DIM incorporates the basic traffic rules and multivariate vehicle motion information. To further improve the prediction accuracy and realize uncertainty estimation, we develop a Gaussian process (GP)-based TPM, considering both the short-term prediction results of the vehicle model and the driving motion characteristics. Afterward, the effectiveness of our novel approach is demonstrated by conducting experiments on a public naturalistic driving dataset under lane-changing scenarios. The superior performance on the task of long-term trajectory prediction is presented and verified by comparing with other advanced methods.

Keywords: Autonomous driving     Dynamic Bayesian network     Driving intention recognition     Gaussian process     Vehicle trajectory prediction    

Thoughts and Suggestions on Autonomous Driving Map Policy

Liu Jingnan, Dong Yang, Zhan Jiao, Gao Kefu

Strategic Study of CAE 2019, Volume 21, Issue 3,   Pages 92-97 doi: 10.15302/J-SSCAE-2019.03.004

Abstract:

As a key infrastructure to realize autonomous driving, autonomous driving map is crucial to the commercial development of the autonomous driving field in China. Currently, constrained by laws and regulations related to ground mapping, map application, and supervision, the commercialization of autonomous driving maps in China is lagging behind. This paper focuses on the main policy and regulatory issues faced in the development, application, and management of autonomous driving maps in China, i.e., encryption of autonomous driving maps, limitations on geographic information expression, qualifications for geographic information collection and the process for map review, accident liability and insurance issues, as well as autonomous driving map related test specifications and test scenario issues. Meanwhile, combining the development trends of domestic and international autonomous driving fields, this paper proposes four suggestions for accelerating the development and commercialization of autonomous driving vehicles in China: formulating an autonomous driving map management mode, allowing pilot application and orderly opening of autonomous driving maps, appropriately opening up corporate authorization and optimizing the review process, as well as establishing a national-level autonomous driving map platform.

Keywords: autonomous driving map     autonomous driving regulation     autonomous driving policy    

A self-supervised method for treatment recommendation in sepsis Research Articles

Sihan Zhu, Jian Pu,jianpu@fudan.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 7,   Pages 926-939 doi: 10.1631/FITEE.2000127

Abstract: treatment is a highly challenging effort to reduce mortality in hospital intensive care units since the treatment response may vary for each patient. Tailored s are desired to assist doctors in making decisions efficiently and accurately. In this work, we apply a self-supervised method based on (RL) for on individuals. An uncertainty evaluation method is proposed to separate patient samples into two domains according to their responses to treatments and the state value of the chosen policy. Examples of two domains are then reconstructed with an auxiliary transfer learning task. A distillation method of privilege learning is tied to a variational auto-encoder framework for the transfer learning task between the low- and high-quality domains. Combined with the self-supervised way for better state and action representations, we propose a deep RL method called high-risk uncertainty (HRU) control to provide flexibility on the trade-off between the effectiveness and accuracy of ambiguous samples and to reduce the expected mortality. Experiments on the large-scale publicly available real-world dataset MIMIC-III demonstrate that our model reduces the estimated mortality rate by up to 2.3% in total, and that the estimated mortality rate in the majority of cases is reduced to 9.5%.

Keywords: 治疗推荐;脓毒症;自监督学习;强化学习;电子病历    

Fully Self-driving Future Hits the Brakes

Chris Palmer

Engineering 2023, Volume 26, Issue 7,   Pages 6-8 doi: 10.1016/j.eng.2023.05.002

Representation learning via a semi-supervised stacked distance autoencoder for image classification Research Articles

Liang Hou, Xiao-yi Luo, Zi-yang Wang, Jun Liang,jliang@zju.edu.cn

Frontiers of Information Technology & Electronic Engineering 2020, Volume 21, Issue 7,   Pages 963-1118 doi: 10.1631/FITEE.1900116

Abstract: is an important application of deep learning. In a typical classification task, the classification accuracy is strongly related to the features that are extracted via deep learning methods. An is a special type of , often used for dimensionality reduction and feature extraction. The proposed method is based on the traditional , incorporating the “distance” information between samples from different categories. The model is called a semi-supervised distance . Each layer is first pre-trained in an unsupervised manner. In the subsequent supervised training, the optimized parameters are set as the initial values. To obtain more suitable features, we use a stacked model to replace the basic structure with a single hidden layer. A series of experiments are carried out to test the performance of different models on several datasets, including the MNIST dataset, street view house numbers (SVHN) dataset, German traffic sign recognition benchmark (GTSRB), and CIFAR-10 dataset. The proposed semi-supervised distance method is compared with the traditional , sparse , and supervised . Experimental results verify the effectiveness of the proposed model.

Keywords: 自动编码器;图像分类;半监督学习;神经网络    

Learning to select pseudo labels: a semi-supervised method for named entity recognition Research Articles

Zhen-zhen Li, Da-wei Feng, Dong-sheng Li, Xi-cheng Lu,lizhenzhen14@nudt.edu.cn,davyfeng.c@gmail.com,dsli@nudt.edu.cn,xclu@nudt.edu.cn

Frontiers of Information Technology & Electronic Engineering 2020, Volume 21, Issue 6,   Pages 809-962 doi: 10.1631/FITEE.1800743

Abstract: models have achieved state-of-the-art performance in (NER); the good performance, however, relies heavily on substantial amounts of labeled data. In some specific areas such as medical, financial, and military domains, labeled data is very scarce, while is readily available. Previous studies have used to enrich word representations, but a large amount of entity information in is neglected, which may be beneficial to the NER task. In this study, we propose a for NER tasks, which learns to create high-quality labeled data by applying a pre-trained module to filter out erroneous pseudo labels. Pseudo labels are automatically generated for and used as if they were true labels. Our semi-supervised framework includes three steps: constructing an optimal single neural model for a specific NER task, learning a module that evaluates pseudo labels, and creating new labeled data and improving the NER model iteratively. Experimental results on two English NER tasks and one Chinese clinical NER task demonstrate that our method further improves the performance of the best single neural model. Even when we use only pre-trained static word embeddings and do not rely on any external knowledge, our method achieves comparable performance to those state-of-the-art models on the CoNLL-2003 and OntoNotes 5.0 English NER tasks.

Keywords: 命名实体识别;无标注数据;深度学习;半监督学习方法    

Autonomous Driving in the iCity—HD Maps as a Key Challenge of the Automotive Industry Perspective

Heiko G. Seif, Xiaolong Hu

Engineering 2016, Volume 2, Issue 2,   Pages 159-162 doi: 10.1016/J.ENG.2016.02.010

Abstract:

This article provides in-depth insights into the necessary technologies for automated driving in future cities. State of science is reflected from different perspectives such as in-car computing and data management, road side infrastructure, and cloud solutions. Especially the challenges for the application of HD maps as core technology for automated driving are depicted in this article.

Keywords: Autonomous driving     Traffic infrastructure     iCity     Car-to-X communication     Connected vehicle     HD maps    

A Hardware Platform Framework for an Intelligent Vehicle Based on a Driving Brain Article

Deyi Li,Hongbo Gao

Engineering 2018, Volume 4, Issue 4,   Pages 464-470 doi: 10.1016/j.eng.2018.07.015

Abstract:

The type, model, quantity, and location of sensors installed on the intelligent vehicle test platform are different, resulting in different sensor information processing modules. The driving map used in intelligent vehicle test platform has no uniform standard, which leads to different granularity of driving map information. The sensor information processing module is directly associated with the driving map information and decision-making module, which leads to the interface of intelligent driving system software module has no uniform standard. Based on the software and hardware architecture of intelligent vehicle, the sensor information and driving map information are processed by using the formal language of driving cognition to form a driving situation graph cluster and output to a decision-making module, and the output result of the decision-making module is shown as a cognitive arrow cluster, so that the whole process of intelligent driving from perception to decision-making is completed. The formalization of driving cognition reduces the influence of sensor type, model, quantity, and location on the whole software architecture, which makes the software architecture portable on different intelligent driving hardware platforms.

Keywords: Driving brain     Intelligent driving     Hardware platform framework    

Interactive medical image segmentation with self-adaptive confidence calibration

沈楚云,李文浩,徐琪森,胡斌,金博,蔡海滨,朱凤平,李郁欣,王祥丰

Frontiers of Information Technology & Electronic Engineering 2023, Volume 24, Issue 9,   Pages 1332-1348 doi: 10.1631/FITEE.2200299

Abstract: Interactive medical image segmentation based on human-in-the-loop machine learning is a novel paradigm that draws on human expert knowledge to assist medical image segmentation. However, existing methods often fall into what we call interactive misunderstanding, the essence of which is the dilemma in trading off short- and long-term interaction information. To better use the interaction information at various timescales, we propose an interactive segmentation framework, called interactive MEdical image segmentation with self-adaptive Confidence CAlibration (MECCA), which combines action-based confidence learning and multi-agent reinforcement learning. A novel confidence network is learned by predicting the alignment level of the action with short-term interaction information. A confidence-based reward-shaping mechanism is then proposed to explicitly incorporate confidence in the policy gradient calculation, thus directly correcting the model’s interactive misunderstanding. MECCA also enables user-friendly interactions by reducing the interaction intensity and difficulty via label generation and interaction guidance, respectively. Numerical experiments on different segmentation tasks show that MECCA can significantly improve short- and long-term interaction information utilization efficiency with remarkably fewer labeled samples. The demo video is available at https://bit.ly/mecca-demo-video.

Keywords: Medical image segmentation     Interactive segmentation     Multi-agent reinforcement learning     Confidence learning     Semi-supervised learning    

NGAT: attention in breadth and depth exploration for semi-supervised graph representation learning Research Articles

Jianke HU, Yin ZHANG,yinzh@zju.edu.cn

Frontiers of Information Technology & Electronic Engineering 2022, Volume 23, Issue 3,   Pages 409-421 doi: 10.1631/FITEE.2000657

Abstract: Recently, graph neural networks (GNNs) have achieved remarkable performance in representation learning on graph-structured data. However, as the number of network layers increases, GNNs based on the neighborhood aggregation strategy deteriorate due to the problem of oversmoothing, which is the major bottleneck for applying GNNs to real-world graphs. Many efforts have been made to improve the process of feature information aggregation from directly connected nodes, i.e., breadth exploration. However, these models perform the best only in the case of three or fewer layers, and the performance drops rapidly for deep layers. To alleviate oversmoothing, we propose a nested graph network (NGAT), which can work in a semi-supervised manner. In addition to breadth exploration, a -layer NGAT uses a layer-wise aggregation strategy guided by the mechanism to selectively leverage feature information from the -order neighborhood, i.e., depth exploration. Even with a 10-layer or deeper architecture, NGAT can balance the need for preserving the locality (including root node features and the local structure) and aggregating the information from a large neighborhood. In a number of experiments on standard tasks, NGAT outperforms other novel models and achieves state-of-the-art performance.

Keywords: Graph learning     Semi-supervised learning     Node classification     Attention    

Title Author Date Type Operation

Pre-training with asynchronous supervised learning for reinforcement learning based autonomous driving

Yunpeng Wang, Kunxian Zheng, Daxin Tian, Xuting Duan, Jianshan Zhou,ypwang@buaa.edu.cn,zhengkunxian@buaa.edu.cn,dtian@buaa.edu.cn,duanxuting@buaa.edu.cn

Journal Article

Toward Human-in-the-loop AI: Enhancing Deep Reinforcement Learning Via Real-time Human Guidance for Autonomous Driving

Jingda Wu, Zhiyu Huang, Zhongxu Hu, Chen Lv

Journal Article

Stochastic pedestrian avoidance for autonomous vehicles using hybrid reinforcement learning

Huiqian LI, Jin HUANG, Zhong CAO, Diange YANG, Zhihua ZHONG,lihq20@mails.tsinghua.edu.cn,huangjin@tsinghua.edu.cn,caoc15@mails.tsinghua.edu.cn,ydg@tsinghua.edu.cn

Journal Article

Towards the Unified Principles for Level 5 Autonomous Vehicles

Jianqiang Wang, Heye Huang, Keqiang Li, Jun Li

Journal Article

Embedding expert demonstrations into clustering buffer for effective deep reinforcement learning

Shihmin WANG, Binqi ZHAO, Zhengfeng ZHANG, Junping ZHANG, Jian PU

Journal Article

A Probabilistic Architecture of Long-Term Vehicle Trajectory Prediction for Autonomous Driving

Jinxin Liu, Yugong Luo, Zhihua Zhong, Keqiang Li, Heye Huang, Hui Xiong

Journal Article

Thoughts and Suggestions on Autonomous Driving Map Policy

Liu Jingnan, Dong Yang, Zhan Jiao, Gao Kefu

Journal Article

A self-supervised method for treatment recommendation in sepsis

Sihan Zhu, Jian Pu,jianpu@fudan.edu.cn

Journal Article

Fully Self-driving Future Hits the Brakes

Chris Palmer

Journal Article

Representation learning via a semi-supervised stacked distance autoencoder for image classification

Liang Hou, Xiao-yi Luo, Zi-yang Wang, Jun Liang,jliang@zju.edu.cn

Journal Article

Learning to select pseudo labels: a semi-supervised method for named entity recognition

Zhen-zhen Li, Da-wei Feng, Dong-sheng Li, Xi-cheng Lu,lizhenzhen14@nudt.edu.cn,davyfeng.c@gmail.com,dsli@nudt.edu.cn,xclu@nudt.edu.cn

Journal Article

Autonomous Driving in the iCity—HD Maps as a Key Challenge of the Automotive Industry

Heiko G. Seif, Xiaolong Hu

Journal Article

A Hardware Platform Framework for an Intelligent Vehicle Based on a Driving Brain

Deyi Li,Hongbo Gao

Journal Article

Interactive medical image segmentation with self-adaptive confidence calibration

沈楚云,李文浩,徐琪森,胡斌,金博,蔡海滨,朱凤平,李郁欣,王祥丰

Journal Article

NGAT: attention in breadth and depth exploration for semi-supervised graph representation learning

Jianke HU, Yin ZHANG,yinzh@zju.edu.cn

Journal Article