人工智能 | CCF推荐期刊专刊约稿信息6条

人工智能

Neurocomputing

Special Issue on INTELLIGENT SENSING, NEURAL COMPUTING AND APPLICATIONS

全文截稿: 2019-05-15

影响因子: 3.241

CCF分类: C类

中科院JCR分区:

  • 大类 : 工程技术 - 2区

  • 小类 : 计算机：人工智能 - 2区

网址: http://www.journals.elsevier.com/neurocomputing/

With the rapid development of sensing and computing technologies, the amount of data in the Internet and in our daily life with different forms is experiencing a rapid growth. Cutting-edge technologies are required to deal with these vast and diverse data streams. Generally, we need to solve some challenges, e.g., how to process, analyze or integrate the data, how to extract useful features from data, how to retrieve knowledge from big data, etc, such that we can explore useful information from this data. Computational intelligence, including neural networks, evolutionary computing, fuzzy logic, etc., receives much attention in industrial and academic communities due to its strong ability for data processing.

In recent years, due to the wide success achieved by deep learning models, neural networks have attracted lots of research interest. A set of representative neural network models have been developed, such as radial basis function neural network, wavelet-based hybrid neural networks, deep boltzmann machine, deep auto-encoder, convolutional neural networks, recurrent neural networks, and generative adversarial networks. And these neural network models have been extensively applied to a wide range of fields, e.g., pattern recognition, data mining, decision making, etc. With advances in computing and networking technologies, more complicated and advanced forms of neural networks are expected to emerge, requiring the design of efficient learning algorithms.

The primary purpose of this special issue is to organise a collection of recently developed neural network theories, methodologies and the applications such as classification, multi-source data computing, learning systems, and decision-making activities. The special issue is intended to be an international forum for researchers to report the recent developments in these fields in an original research paper style.

人工智能

Neurocomputing

Special Issue on Privacy-preserving Representation Learning for Big Data (PRLBD)

全文截稿: 2019-05-31

影响因子: 3.241

CCF分类: C类

中科院JCR分区:

  • 大类 : 工程技术 - 2区

  • 小类 : 计算机：人工智能 - 2区

网址: http://www.journals.elsevier.com/neurocomputing/

Big data needs huge storage space, and its applications require powerful computation capability. Recently, it is very popular for data owners to outsource big data from local servers to cloud due to the great flexibility and cost saving of cloud computing, such as managing the massive computation workload in representation learning and data retrieval. However, representation learning on cloud data may reveal privacy of data owners, such as personal identity, location, and financial profiles hobbies. To do this, data owners can encrypt their data for confidentiality before uploading them to cloud. However, encrypted data makes its feature extraction (i.e., representation learning) difficult. The attacker can deduce the data content via feature comparisons in benchmark data sets, or even recover part of the data based on the derived features. Thus, exploring privacy-preserving representation learning from big data becomes of very importance in the domains of both machine learning and data cybersecurity.

In this special issue, we invite papers to address many challenges of representation learning from big data. EspeciallySpecifically, to provide readers of the this special issue with a state-of-the-art background on the topic, we will invite one survey paper, which will undergo the peer review process. The list of possible topics include, but not limited to:

Privacy-preserving representation learning for big multimedia data

Data preprocess including missing value imputation, feature selection , clustering, and synthesizing/fusion

Distributed/paralleled/sampling techniques for big data

Sparse/Dictionary learning from big data

Privacy-preserving representation learning or feature extraction

Privacy-preserving deep learning

Privacy-preserving deep learning framework

Secured kNN algorithm on encrypted data

Secured deep hashing techniques for image retrieval

Secured representation learning methods using deep learning models

Privacy-preserving representation learning for multi-task data

Secured similarities/dissimilarities learning from multiple tasks

Secured regularization strategies in multi-task learning

Large tasks (modals), small sample size learning for secured multi-task learning, domain adaptation, and transfer learning

人工智能

Pattern Recognition

Call for papers of a special issue on Deep Video Analysis: Models, Algorithms and Applications

全文截稿: 2019-05-31

影响因子: 3.962

CCF分类: B类

中科院JCR分区:

  • 大类 : 工程技术 - 2区

  • 小类 : 计算机：人工智能 - 2区

  • 小类 : 工程：电子与电气 - 2区

网址: http://www.journals.elsevier.com/pattern-recognition/

Video analysis is an important research area in pattern recognition and computer vision. The past decades have witnessed the rapid expansion of the video data generated every day including video surveillance, personal mobile device capture, and webs upload. It is quite needed for understanding such a large amount of video data. Meanwhile, deep learning, as a fast-growing research field, showed a vast successover a wide of research areas. The combination of the big visual data and the deep learning paradigm would bring a significant progress in both video analysis and life-related applications. However, the current capability gap of deep learning to handle video analysis is still huge. This mainly because video data conveys rich spatial and temporal information. In addition, the annotation of video labels especially pixel-level labels is expensively-acquired, which limits the further learning of deep neural networks. Thus, there is a pressing demand for novel deep learning based video analysis approaches that can cope with video analysis task with better accuracy and efficiency. In this special issue, we attempt to assemble recent advances in the deep learning based video analysis and related extended applications.

This special issue will feature original research papers related to the models and algorithm for various video analysis tasks from the low-level (e.g., video processing) to high-level (e.g., spatiotemporal reasoning), together with widespread applications to real-world issues. The main topics of interest (but are not limited to):

— Learning data representation from video based on supervised/unsupervised/semi-supervised learning.

—Video object tracking and segmentation: 1) Single object/multiple objects tracking; 2) Video segmentation; 3) Object saliency prediction.

—Optical flow estimation and visual SLAM.

—Person re-identification, multiple-camera tracking and vehicle re-identification, and human pose estimation/tracking.

—Dynamic scene parsing and semantic video segmentation.

—Video understanding: 1) Video classification and spatiotemporal reasoning; 2) Video action recognition; 3) Video summarization/video captioning; 4) Video generation; 5) Video anomaly detection.

—Applications: Applications of the corresponding methods to solve real-world video analysis issues including robot visions, machine visions, video object detection etc.

—New benchmark datasets related to the aforementioned topics.

人工智能

Neurocomputing

Special Issue on Advances in Deep and Shallow Machine Learning Approaches for Handling Data Irregularities

全文截稿: 2019-06-15

影响因子: 3.241

CCF分类: C类

中科院JCR分区:

  • 大类 : 工程技术 - 2区

  • 小类 : 计算机：人工智能 - 2区

网址: http://www.journals.elsevier.com/neurocomputing/

Performance of most of the well-known learning systems can considerably degrade if the data to be handled (e.g. the training examples for supervised learning) contain irregularities of various types. By data irregularity, we point to those situations where the distribution of data points, the sampling of data space for generating the training set, or the features describing each data point deviate from what could have been ideal; being biased, skewed, incomplete and/or misleading. Data irregularities can give rise to problems of class imbalance, small disjuncts (rarely covered sub-concepts within each class), class distribution skew, absent features, and missing features. While the three former problems are defined only for supervised learning systems, the latter two are defined for supervised as well as unsupervised learning systems. More than one of these assumptions may be violated together by a given dataset. Majority of the learning systems (both supervised and unsupervised) are quite sensitive to one of more of the data irregularities mentioned above, irrespectively of the systems?generalization ability or ability to learn representations. Unless special mechanisms are adopted either for data pre-processing or as a part of the learning system itself, such data irregularities considerably degrade the performance of traditional computer-based learning systems. This special issue aims to report the latest advances in the design of learning systems which can show considerable resilience to data irregularities of various forms which are frequently present in most of the real-life data.

The topics include, but are not limited to:

Handling class imbalance in supervised learning

Handling small disjuncts and sub-concepts within classes.

Handling class distribution skew and handling the disparate shape of class distributions

Missing and absent features problem in classification and clustering

Label noise and ambiguity resolution in supervised learning

Interrelation and co-occurrence of data irregularities

Effect of data irregularities on deep neural networks

Effect of data irregularities on adversarial machine learning

Theoretical studies on data preprocessing and learning algorithms tailor-made to handle data irregularities

New application areas giving rise to data irregularities and design of relevant learning systems including object detection, classification, recognition; image retrieval, segmentation, interpretation; document retrieval, categorization, topic model; signal processing, medical image analysis, bioinformatics; speech recognition, synthesis, anti-spoofing; fault detection/diagnosis, fraud detection, cyber-security; etc.

人工智能

Neurocomputing

Special Issue on Deep Learning for Human Activity Recognition

全文截稿: 2019-09-01

影响因子: 3.241

CCF分类: C类

中科院JCR分区:

  • 大类 : 工程技术 - 2区

  • 小类 : 计算机：人工智能 - 2区

网址: http://www.journals.elsevier.com/neurocomputing/

Human activity recognition (HAR) can benefit various applications, such as health-care services and smart home applications. Many sensors have been utilized for human activity recognition, such as wearable sensors, smartphones, radio frequency (RF) sensors (WiFi, RFID), LED light sensors, cameras, etc. Owing to the rapid development of wireless sensor network, a large amount of data has been collected for the recognition of human activities with different kind of sensors. Conventional shallow learning algorithms, such as support vector machine and random forest, require to manually extract some representative features from large and noisy sensory data. However, manual feature engineering requires export knowledge and will inevitably miss implicit features.

Recently, deep learning has achieved great success in many challenging research areas, such as image recognition and natural language processing. The key merit of deep learning is to automatically learn representative features from massive data. This technology can be a good candidate for human activity recognition. Some initial attempts can be found in the literature. However, many challenging research problems in terms of accuracy, device heterogeneous, environment changes, etc. remain unsolved.

This special collection intends to prompt state-of-the-art approaches on deep learning for human activity recognition. All submitted papers will be peer-reviewed and selected based on both their quality and relevance.

Potential topics include but are not limited to:

Device-based HAR using deep learning

Device-free HAR using deep learning

Image based HAR using deep learning

Light sensor based HAR using deep learning

Sensor fusion for HAR using deep learning

Fusion of shallow models with deep networks for HAR

Device heterogeneous for device-based HAR

Environment changes for device-free HAR

Transfer Learning for HAR

Online Learning for HAR

Semi-supervised Learning for HAR

Survey for deep learning based HAR

人工智能

Pattern Recognition Letters

Special Issue on Artificial Intelligence for Distributed Smart Sensing

全文截稿: 2019-12-31

影响因子: 1.952

CCF分类: C类

中科院JCR分区:

  • 大类 : 工程技术 - 3区

  • 小类 : 计算机：人工智能 - 3区

网址: http://www.journals.elsevier.com/pattern-recognition-letters/

The goal of Artificial Intelligence (AI) is to reproduce biological intelligence in the form of adaptive machines.

The path towards this goal is characterized by several steps, among which the integration of the AI with Smart Sensors (SS) is fundamental. SS and, more generally, Smart Cyber Physical Systems are nowadays significantly impacting the everyday life of citizens and, in perspective, they will become pervasive in every aspect of human life from public health and well-being to home, infrastructures and environment management.

It is only thanks to the integration of AI and SS that computers can increasingly see, hear, touch, smell and taste and so become aware and capable to positively interact with the environment in which they are deployed.

The research activity (industrial and scientific) in AI is still very fragmented. In fact the development of an intelligent system capable of dealing with all the senses and adapting to different contexts is still relatively far. Furthermore, the results obtained on different sensing areas are still very unbalanced.

Indeed, the obtained results are impressive for some senses and weak for the others. Into the first category it is possible to include sight (with vision systems made by large companies and research institutes), hearing (with the speech to text systems of many devices for everyday use) and the more general "comprehension".

For the other senses, much more work remains to be done: touch sensors are little more than devices able to understand if "I’m touching something", whereas on smell and taste there is still much to be done.

Another important issue is related to the possibility of exploiting collaborative approaches through Distributed Architectures. In this kind of applications, SS are spread into the environment of interest where some kind of “social intelligence” is generated. Many applications of such an architecture are possible in smart cities, smart industries, smart buildings, etc.

The improvements will necessarily have to take place at different levels: physical (sensors with increased discriminatory capabilities, robustness and stability), data processing (sensors equipped with electronics for signal conditioning in order to make them "informative"), data communication (sensors equipped with different solutions for sending/receiving data following for example the IoT paradigm) and, finally, understanding the data (with AI).

The aim of this Special Issue is to bring together academics and industrial practitioners to exchange and discuss the latest innovations and applications of AI in the domain of SS and DSS.

TOPICS

- Wired and wireless solutions

- New sensor technologies

- Internet of Things

- Computer Vision

- Natural Language processing

- Deep and Reinforcement Learning

- Ontology solutions

- Sensor Network

- Critical applications

- Soft Computing

- Computational Intelligence

- Neurocomputing/Neural Systems

- Case studies

- Multimedia Learning

- Classification and clustering algorithms for DSS.

- Solutions for Industry 4.0 (energy, logistics, optimization, ...)

- Agent-based solutions

- Wearable solutions

- Big data analysis

- Ambient assisted living

- Hazard detection

- Real Time

- Security

- Applications of AI

- Multi-modal distributed sensors

- Distributed sensing applications

all referred to SS and/or to DSS.

下载Call4Papers App，获取更多详细内容！