Soft-computing remains as one of the promising disciplines that successfully address many different real-world problems. In such circumstances, where demanding constraints are present, this set of ...

Novel solutions, based on soft-computing techniques, are proposed in the present issue. All of them target open problems in the environmental and industrial domains. Thanks to the intelligent syste...

This special issue compiles recent applications of Computational Intelligence in the security domain. It is aimed at both researchers and practitioners from academia and industry who are engaged in deploying intelligent solutions for securing both information and systems. Four papers are included in this special issue, covering a wide variety of case studies, ranging from assisted driving to explainable robotics. Additionally, a variety of cutting-edge techniques are proposed, such as blockchain and autoencoders among others. All these works contribute to the scientific progress in this domain, going one step further. In the first contribution by Barreno et al., it is proposed a fuzzy expert system aimed at identifying and classifying conventional two-lane roads based on their geometric characteristics and the expected vehicle performance (passenger cars, trucks, and buses). The road geometric features were measured by sensors in an equipped vehicle, travelling in real-life roads located in the Madrid region (Spain). Fuzzy and neuro-fuzzy systems are applied in order to classify road sections based on their current geometric characteristics. Additionally, a risk identification system is presented. Its main target is to assess whether a vehicle is driving on a two-lane road with an inappropriate speed. In order to do that, an identification model is applied to some variables such as the vehicle type, the road longitudinal gradient, the angle covered by each horizontal curve, and the existence or not of an additional lane of traffic. A fuzzy risk index is proposed to verify relatively unsafe road sections, where more accidents are prone to happen. The next contribution, by Wilusz and Wojtowicz, proposes an architecture for cryptocurrency insurance framework along with a set of corresponding communication protocols and a smart contract template that jointly reduce various risks to which cryptocurrency-based transactions are exposed. Some of the threats are unintentional credential leaks, fraudulent cryptocurrency exchanges and broken private keys among others. Authors analyse such risks and present a solution that can potentially enable new business models based on this technology. In the proposed insurance framework, the holder transfers the risk to the insurer and the insurer applies technological countermeasures to secure the cryptocurrency unit. Funds can be transferred with the bilateral agreement of the cryptocurrency unit holder and the insurer thanks to the use of multi-signatures required by smart contracts. This framework is analysed from the security standpoint, guaranteeing that it can be deployed in several practical scenarios in various domains of applications. Zayas-Gato et al. address the challenging and up-to-date issue of detecting anomalies in industrial systems. In order to automatically detect the anomalies, a hybrid system is proposed. First, data are clustered by the Density-Based Spatial Clustering of Applications With Noise (DBSCAN) algorithm. Then, eight one-class techniques are benchmarked to model the normal operation the target industrial plants, namely: NCBoP, Autoencoders, Gauss, K-Centers, Minimum Spanning Trees, Parzen Density Estimator, Principal Component Analysis, and Support Vector Data Description. Authors' proposal is validated using two real-life domains with data gathered during different operating points of two industrial plants. The first case is controlling the liquid level in a tank while the second one is manufacturing wind generator blades made of carbon fibre material. The validated approach can be used to improve the classification when the target set is scattered in different clusters or groups. As a result, an early detection of anomalous situations is more feasible, increasing the systems optimisation. The last contribution, by Rodríguez-Lera et al., focus on accountability of intelligent robots. This implies that any robot is in charge of logging its activities with verifiable evidence so that all actions are traceable and the events triggering the action are identified. The main contributions of this study are an overview, modelling, and formalization of the three perspectives of accountability for autonomous robots; a GUI tool to graphically illustrate the knowledge supported on Conceptual Graphs; and finally, a proof of concept summarizing the information dumped on each level using a set of innovative debugging tools. The findings of this research support the idea that a verbose logging system drives to serious performance issues on an accountability platform, and the detailed information provided is far from explaining the robot's behaviour. Besides, large volumes of data might not be managed properly in specific situations; also, the access to such information sacrifices the privacy of the human-robot interaction. The target is that the robot could incorporate the mechanisms that help manufacturers, client/user and developers to know the reasons that trigger a certain behaviour. The guest editors wish to thank Prof. Jon G. Hall (Editor-in-Chief of the Wiley-Blackwell Journal Expert Systems: The Journal of Knowledge Engineering) for providing the opportunity to edit this special issue. Additional thanks to Prof. Lucia Rapanotti, for her support in managing the editing process. Finally, guest editors would also like to thank the referees who have thoroughly evaluated the papers and the editorial staff for their support. The authors declare no potential conflict of interest.

Abstract Material flow management refers to the analysis and specific optimization of the inventory‐production system. Material flow can be characterized as the organized flow of material in a production process with the required sequence determined by a technological procedure. The Milk run system assures the transportation of materials at the right time and in an optimal manner. It should be combined with the Kanban system to highlight when something is required in the production process. This paper presents biological swarm intelligence, in general, and a particular model, particle swarm optimization (PSO), for modelling material flow using a Milk run system supported by a Kanban system in the automotive industry. The aim of this study is to create a new model for the optimal number of trailers for one train and optimal number of containers in a tugger train system when the route time period has been defined. A new modified PSO approach for integrating inventory‐production in a unique optimization model is used. The major modification to the original PSO is using the capacity of a container instead of a velocity component. Each new Kanban trigger is checked, and the total timing for the Milk run delivery solution is calculated for the necessary raw material capacity for each shop floor.

"Recent Advances on the Application of Soft Computing to Industrial and Environmental Enterprises." Cybernetics and Systems, 51(7), pp. 647–648

A feedback solution for approximate optimal scheduling of switched systems with autonomous subsystems and continuous-time dynamics is presented. The proposed solution is based on policy iteration algorithm which provides the optimal switching schedule. Algorithms for offline, online, and concurrent implementation of the proposed solution are presented. For online and concurrent training, gradient descent training laws are used and the performance of the training laws is analyzed. The effectiveness of the presented algorithms is verified through numerical simulations.

Current advances in the development of mobile and smart devices have generated a growing demand for natural human-machine interaction and favored the intelligent assistant metaphor, in which a single interface gives access to a wide range of functionalities and services. Conversational systems constitute an important enabling technology in this paradigm. However, they are usually defined to interact in semantic-restricted domains in which users are offered a limited number of options and functionalities. The design of multi-domain systems implies that a single conversational system is able to assist the user in a variety of tasks. In this paper we propose an architecture for the development of multi-domain conversational systems that allows: (1) integrating available multi and single domain speech recognition and understanding modules, (2) combining available system in the different domains implied so that it is not necessary to generate new expensive resources for the multi-domain system, (3) achieving better domain recognition rates to select the appropriate interaction management strategies. We have evaluated our proposal combining three systems in different domains to show that the proposed architecture can satisfactory deal with multi-domain dialogs.

In this research, a novel family of learning rules called Beta Hebbian Learning (BHL) is thoroughly investigated to extract information from high-dimensional datasets by projecting the data onto low-dimensional (typically two dimensional) subspaces, improving the existing exploratory methods by providing a clear representation of data's internal structure. BHL applies a family of learning rules derived from the Probability Density Function (PDF) of the residual based on the beta distribution. This family of rules may be called Hebbian in that all use a simple multiplication of the output of the neural network with some function of the residuals after feedback. The derived learning rules can be linked to an adaptive form of Exploratory Projection Pursuit and with artificial distributions, the networks perform as the theory suggests they should: the use of different learning rules derived from different PDFs allows the identification of "interesting" dimensions (as far from the Gaussian distribution as possible) in high-dimensional datasets. This novel algorithm, BHL, has been tested over seven artificial datasets to study the behavior of BHL parameters, and was later applied successfully over four real datasets, comparing its results, in terms of performance, with other well-known Exploratory and projection models such as Maximum Likelihood Hebbian Learning (MLHL), Locally-Linear Embedding (LLE), Curvilinear Component Analysis (CCA), Isomap and Neural Principal Component Analysis (Neural PCA).

Abstract The unequal area facility layout problem (UA‐FLP) has been addressed by many methods. Most of them only take aspects that can be quantified into account. This contribution presents a novel approach, which considers both quantitative aspects and subjective features. To this end, a multi‐objective interactive genetic algorithm is proposed with the aim of allowing interaction between the algorithm and the human expert designer, normally called the decision maker (DM) in the field of UA‐FLP. The contribution of the DM's knowledge into the approach guides the complex search process, adjusting it to the DM's preferences. The entire population associated to facility layout designs is evaluated by quantitative criteria in combination with an assessment prepared by the DM, who gives a subjective evaluation for a set of representative individuals of the population in each iteration. In order to choose these individuals, a soft computing clustering method is used. Two interesting real‐world data sets are analysed to empirically probe the robustness of these models. The first UA‐FLP case study describes an ovine slaughterhouse plant and the second, a design for recycling carton plant. Relevant results are obtained, and interesting conclusions are drawn from the application of this novel intelligent framework.

Dealing with distributed data is one of the challenges for clustering, as most clustering techniques require the data to be centralized. One of them, k-means, has been elected as one of the most influential data mining algorithms for being simple, scalable, and easily modifiable to a variety of contexts and application domains. However, exact distributed versions of k-means are still sensitive to the selection of the initial cluster prototypes and require the number of clusters to be specified in advance. Additionally, preserving data privacy among repositories may be a complicating factor. In order to overcome k-means limitations, two different approaches were adopted in this paper: the first obtains a final model identical to the centralized version of the clustering algorithm and the second generates and selects clusters for each distributed data subset and combines them afterwards. It is also described how to apply the algorithms compared while preserving data privacy. The algorithms are compared experimentally from two perspectives: the theoretical one, through asymptotic complexity analyses, and the experimental one, through a comparative evaluation of results obtained from a collection of experiments and statistical tests. The results obtained indicate which algorithm is more suitable for each application scenario.

A challenging task in time-course microarray data analysis is to cluster genes meaningfully combining the information provided by multiple replicates covering the same key time points. This paper proposes a novel cluster merging method to accomplish this goal obtaining groups with highly correlated genes. The main idea behind the proposed method is to generate a clustering starting from groups created based on individual temporal series (representing different biological replicates measured in the same time points) and merging them by taking into account the frequency by which two genes are assembled together in each clustering. The gene groups at the level of individual time series are generated using several shape-based clustering methods. This study is focused on a real-world time series microarray task with the aim to find co-expressed genes related to the production and growth of a certain bacteria. The shape-based clustering methods used at the level of individual time series rely on identifying similar gene expression patterns over time which, in some models, are further matched to the pattern of production/growth. The proposed cluster merging method is able to produce meaningful gene groups which can be naturally ranked by the level of agreement on the clustering among individual time series. The list of clusters and genes is further sorted based on the information correlation coefficient and new problem-specific relevant measures. Computational experiments and results of the cluster merging method are analyzed from a biological perspective and further compared with the clustering generated based on the mean value of time series and the same shape-based algorithm.

Accurate intrusion detection is still an open challenge. The present work aims at being one step toward that purpose by studying the combination of clustering and visualization techniques. To do that, the mobile visualization connectionist agent-based intrusion detection system (MOVICAB-IDS), previously proposed as a hybrid intelligent IDS based on visualization techniques, is upgraded by adding automatic response thanks to clustering methods. To check the validity of the proposed clustering extension, it has been applied to the identification of different anomalous situations related to the simple network management network protocol by using real-life data sets. Different ways of applying neural projection and clustering techniques are studied in the present article. Through the experimental validation it is shown that the proposed techniques could be compatible and consequently applied to a continuous network flow for intrusion detection.

Abstract This interdisciplinary research is based on the application of unsupervized connectionist architectures in conjunction with modelling systems and on the determining of the optimal operating conditions of a new high precision industrial process known as laser milling. Laser milling is a relatively new micro‐manufacturing technique in the production of high‐value industrial components. The industrial problem is defined by a data set relayed through standard sensors situated on a laser‐milling centre, which is a machine tool for manufacturing high‐value micro‐moulds, micro‐dies and micro‐tools. The new three‐phase industrial system presented in this study is capable of identifying a model for the laser‐milling process based on low‐order models. The first two steps are based on the use of unsupervized connectionist models. The first step involves the analysis of the data sets that define each case study to identify if they are informative enough or if the experiments have to be performed again. In the second step, a feature selection phase is performed to determine the main variables to be processed in the third step. In this last step, the results of the study provide a model for a laser‐milling procedure based on low‐order models, such as black‐box, in order to approximate the optimal form of the laser‐milling process. The three‐step model has been tested with real data obtained for three different materials: aluminium, cooper and hardened steel. These three materials are used in the manufacture of micro‐moulds, micro‐coolers and micro‐dies, high‐value tools for the medical and automotive industries among others. As the model inputs are standard data provided by the laser‐milling centre, the industrial implementation of the model is immediate. Thus, this study demonstrates how a high precision industrial process can be improved using a combination of artificial intelligence and identification techniques.

Neural intelligent systems can provide a visualization of the network traffic for security staff, in order to reduce the widely known high false-positive rate associated with misuse-based Intrusion Detection Systems (IDSs). Unlike previous work, this study proposes an unsupervised neural models that generate an intuitive visualization of the captured traffic, rather than network statistics. These snapshots of network events are immensely useful for security personnel that monitor network behavior. The system is based on the use of different neural projection and unsupervised methods for the visual inspection of honeypot data, and may be seen as a complementary network security tool that sheds light on internal data structures through visual inspection of the traffic itself. Furthermore, it is intended to facilitate verification and assessment of Snort performance (a well-known and widely-used misuse-based IDS), through the visualization of attack patterns. Empirical verification and comparison of the proposed projection methods are performed in a real domain, where two different case studies are defined and analyzed.

The real-world problem of Urban Bicycles Renting Systems (UBRS) in a city requires the optimization of vehicle routes connecting several bicycle base stations and storage centers. This problem can be modeled as a capacitated Vehicle Routing Problem (VRP) with multiple depots and the simultaneaous need for pickup and delivery at each base station location. Based on the VRP model specification, an evolutionary approach is proposed to address the UBRS problem. Individuals are encoded as permutations of base stations and then translated to a set of routes subject to the constraints related to vehicle capacity and node demands. Better-fitted offspring generated via order crossover or swap mutation are asynchronously inserted in the population. The proposed evolutionary algorithm is engaged for the UBRS problem using data from the city of Barcelona with promising results. Some relevant parameters that can enhance the proposed approach have been identified and analysed via the computational experiments.

In recent years, many solutions have been developed for the Nurse Rostering Problem. When a new problem of this kind arrives, it is not so easy to choose the proper solution from previous work. This study presents an application of a bio-inspired intelligent system to analyse and select previous nurse rostering solutions. This applied research presents a multidisciplinary study based on the application of unsupervised neural projection models in order to identify the similar solutions to the mentioned problem. The system has been tested under a real data set gathered from the current state of the art, achieving promising results.

This paper presents a novel model for performing classification and visualization of high-dimensional data by means of combining two enhancing techniques. The first is a semi-supervised learning, an extension of the supervised learning used to incorporate unlabeled information to the learning process. The second is an ensemble learning to replicate the analysis performed, followed by a fusion mechanism that yields as a combined result of previously performed analysis in order to improve the result of a single model. The proposed learning schema, termed S(2)-Ensemble, is applied to several unsupervised learning algorithms within the family of topology maps, such as the Self-Organizing Maps and the Neural Gas. This study also includes a thorough research of the characteristics of these novel schemes, by means quality measures, which allow a complete analysis of the resultant classifiers from the viewpoint of various perspectives over the different ways that these classifiers are used. The study conducts empirical evaluations and comparisons on various real-world datasets from the UCI repository, which exhibit different characteristics, so to enable an extensive selection of situations where the presented new algorithms can be applied.

During the last years there has been a growing need of developing innovative tools that can help small to medium sized enterprises to predict business failure as well as financial crisis. In this study we present a novel hybrid intelligent system aimed at monitoring the modus operandi of the companies and predicting possible failures. This system is implemented by means of a neural-based multi-agent system that models the different actors of the companies as agents. The core of the multi-agent system is a type of agent that incorporates a case-based reasoning system and automates the business control process and failure prediction. The stages of the case-based reasoning system are implemented by means of web services: the retrieval stage uses an innovative weighted voting summarization of self-organizing maps ensembles-based method and the reuse stage is implemented by means of a radial basis function neural network. An initial prototype was developed and the results obtained related to small and medium enterprises in a real scenario are presented.

The significance that Ambient Intelligence (AmI) has acquired in recent years requires the development of innovative solutions. Nonetheless, the development of AmI-based systems requires the creation of increasingly complex and flexible applications. In this regard, the use of context-aware technologies is an essential aspect in these developments to perceive stimuli from the context and react upon it autonomously. This work presents a novel platform that defines a method for integrating dynamic and self-adaptable heterogeneous Wireless Sensor Networks (WSNs). This approach facilitates the inclusion of context-aware capabilities when developing intelligent ubiquitous systems, where functionalities can communicate in a distributed way. Furthermore, the information obtained must be managed by intelligent and self-adaptable technologies to provide an adequate interaction between the users and their environment. Agents and Multi-Agent Systems are one of these technologies. The agents have characteristics such as autonomy, reasoning, reactivity, social abilities and pro-activity which make them appropriate for developing dynamic and distributed systems based on AmI. This way, the integration of the platform with a Service-Oriented Multi-Agent architecture is proposed. Finally, conclusions and future work are presented.

From an enterprise point of view, Knowledge Management (KM) enables organizations to capture, share, and apply the collective experience and know-how (knowledge) of their staff. Up to now, little effort has been devoted to apply Artificial Intelligent techniques to KM systems. This paper proposes the application of case-based reasoning, in combination with a neural model, to develop a KM system. This combined approach profiles the KM status of the whole organization and automatically generates proposals, aimed at improving the KM situation of organization units. The system is fed with KM data collected at the organization and unit contexts. The outcome consists of customized solutions for different areas of expertise related to the organization units, once a lack of knowledge in any of those has been identified.

SQL Injection attacks on web applications have become one of the most important information security concerns over the past few years. This paper presents a hybrid approach based on the Adaptive Intelligent Intrusion Detector Agent (AIIDA-SQL) for the detection of those attacks. The AIIDA-SQL agent incorporates a Case-Based Reasoning (CBR) engine which is equipped with learning and adaptation capabilities for the classification of SQL queries and detection of malicious user requests. To carry out the tasks of attack classification and detection, the agent incorporates advanced algorithms in the reasoning cycle stages. Concretely, an innovative classification model based on a mixture of an Artificial Neuronal Network together with a Support Vector Machine is applied in the reuse stage of the CBR cycle. This strategy enables to classify the received SQL queries in a reliable way. Finally, a projection neural technique is incorporated, which notably eases the revision stage carried out by human experts in the case of suspicious queries. The experimental results obtained on a real-traffic case study show that AIIDA-SQL performs remarkably well in practice.

A novel soft computing system to optimize a dental milling process is proposed. The model is based on the initial application of several statistical and projection methods as Principal Component Analysis and Cooperative Maximum Likelihood Hebbian Learning to analyze the structure of the data set and to identify the most relevant variables. Finally, a supervised neural model and identification techniques are applied, in order to model the process and optimize it. In this study a real data set obtained by a dynamic machining center with five axes simultaneously is analyzed to empirically test the novel system in order to optimize the time error.

This study presents a novel, multidisciplinary research project entitled DIPKIP (data acquisition, intelligent processing, knowledge identification and proposal), which is a Knowledge Management (KM) system that profiles the KM status of a company. Qualitative data is fed into the system that allows it not only to assess the KM situation in the company in a straightforward and intuitive manner, but also to propose corrective actions to improve that situation. DIPKIP is based on four separate steps. An initial “Data Acquisition” step, in which key data is captured, is followed by an “Intelligent Processing” step, using neural projection architectures. Subsequently, the “Knowledge Identification” step catalogues the company into three categories, which define a set of possible theoretical strategic knowledge situations: knowledge deficit, partial knowledge deficit, and no knowledge deficit. Finally, a “Proposal” step is performed, in which the “knowledge processes”—creation/acquisition, transference/distribution, and putting into practice/updating—are appraised to arrive at a coherent recommendation. The knowledge updating process (increasing the knowledge held and removing obsolete knowledge) is in itself a novel contribution. DIPKIP may be applied as a decision support system, which, under the supervision of a KM expert, can provide useful and practical proposals to senior management for the improvement of KM, leading to flexibility, cost savings, and greater competitiveness. The research also analyses the future for powerful neural projection models in the emerging field of KM by reviewing a variety of robust unsupervised projection architectures, all of which are used to visualize the intrinsic structure of high‐dimensional data sets. The main projection architecture in this research, known as Cooperative Maximum‐Likelihood Hebbian Learning (CMLHL), manages to capture a degree of KM topological ordering based on the application of cooperative lateral connections. The results of two real‐life case studies in very different industrial sectors corroborated the relevance and viability of the DIPKIP system and the concepts upon which it is founded.

Oil spills represent one of the most destructive environmental disasters. Predicting the possibility of finding oil slicks in a certain area after an oil spill can be critical in reducing environmental risks. The system presented here uses the Case-Based Reasoning (CBR) methodology to forecast the presence or absence of oil slicks in certain open sea areas after an oil spill. CBR is a computational methodology designed to generate solutions to certain problems by analysing previous solutions given to previously solved problems. The proposed CBR system includes a novel network for data classification and retrieval. This type of network, which is constructed by using an algorithm to summarize the results of an ensemble of Self-Organizing Maps, is explained and analysed in the present study. The Weighted Voting Superposition (WeVoS) algorithm mainly aims to achieve the best topographically ordered representation of a dataset in the map. This study shows how the proposed system, called WeVoS-CBR, uses information such as salinity, temperature, pressure, number and area of the slicks, obtained from various satellites to accurately predict the presence of oil slicks in the north-west of the Galician coast, using historical data.

MOVICAB-IDS enables the more interesting projections of a massive traffic data set to be analysed, thereby providing an overview of any possible anomalous situations taking place on a computer network. This IDS responds to the challenges presented by traffic volume and diversity. It is a connectionist agent-based model extended by means of a functional and mobile visualization interface. The IDS is designed to be more flexible, accessible and portable by running on a great variety of applications, including small mobile ones such as PDA’s, mobile phones or embedded devices. Furthermore, its effectiveness has been demonstrated in different tests.

Case-based reasoning systems, in general, and instance-based reasoning systems, in particular, are used more and more in industrial applications nowadays. During the last few years, researchers have been working in the development of techniques to automate the reasoning stages identified in this methodology. This paper presents a method for automating the retrieval stage and indexation of instance-based reasoning systems. This method is based on a modification of a new type of topology preserving map that can be used for scale invariant classification. The scale invariant map is an implementation of the negative feedback network to form a topology-preserving mapping. Maximum/minimum likelihood learning is applied in this paper to the scale invariant map and its possibilities are explored. This method automates the organization of cases and the retrieval stage of case-based reasoning systems. The proposed methodology groups instances with similar structure, identifying clusters automatically in a data set in an unsupervised mode. The method has been successfully used to completely automate the reasoning process of an oceanographic forecasting system and to improve its performance.

We review an extension of Hebbian learning which has been called ε-Insensitive Hebbian Learning (Fyfe and MacDonald, 2001) and derive lateral connections from a probability density function. We use these lateral connections to move outputs towards the mode of the pdf and use the resulting outputs to train the feedforward connections. We show that the resulting network is able to identify a single orientation of bars from a mixture of horizontal and vertical bars.

We consider the difficult problem of identification of independent causes from a mixture of them when these causes interfere with one another in a particular manner: those considered are visual inputs to a neural network system which are created by independent underlying causes which may occlude each other. The prototypical problem in this area is a mixture of horizontal and vertical bars in which each horizontal bar interferes with the representation of each vertical bar and vice versa. Previous researchers have developed artificial neural networks which can identify the individual causes; we seek to go further in that we create artificial neural networks which identify all the horizontal bars from only such a mixture. This task is a necessary precursor to the development of the concept of "horizontal" or "vertical".

This paper reviews an unsupervised artificial neural network that has been shown to perform principal component analysis and a constrained version of the same network that has been shown to perform factor analysis. It is shown that this network, when trained on real video data, finds filters that are both local in time and local in space. It is further shown that the type of movement and environment in these video sequences determines the shape of the filters found. It is then shown that lateral connections derived from finding the mode of a probability density function can be used to form a global ordering of the output responses of the network and that different parameter regimes can be used to differentiate between competing, mutually interfering classes of factors. Indeed, some parameter values can be used to suppress entirely particular classes of factors. The net effect on video data is that specific types of movement can be identified by examining the network's outputs’ responses.

In this paper, we compare two artificial neural network algorithms for performing Exploratory Projection Pursuit, a statistical technique for investigating data by projecting it onto lower dimensional manifolds. The neural networks are extensions of a network which performs Principal Component Analysis. We illustrate the technique on artificial data before applying it to real data.

This paper is an exploration of a straightforward but powerful unsupervised artificial neural network technique for the efficient coding of artificial and real data. Using an energy function based on the Rectified Gaussian distribution, we derive a neural architecture based on an unsupervised negative feedback network with fixed lateral connections. We show that, not only can this network identify local correlated structure in visual data but through the use of appropriate lateral connections, we can obtain a grouping of similar causes on the output response of the network. We show that the network may be used to form local spatiotemporal filters in response to real images contained in video. The shape of these filters reflects the nature of the video sequences.