2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)最新文献

{"title":"Analysis of adaptation law of the robust evolving cloud-based controller","authors":"G. Andonovski, S. Blažič, P. Angelov, I. Škrjanc","doi":"10.1109/EAIS.2015.7368793","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368793","url":null,"abstract":"In this paper we propose a performance analysis of the robust evolving cloud-based controller (RECCo) according to the different initial scenarios. RECCo is a controller based on fuzzy rule-based (FRB) systems with non-parametric antecedent part and PID type consequent part. Moreover, the controller structure (the fuzzy rules and the membership function) is created in online manner from the data stream. The advantage of the RECCo controller is that do not require any a priory knowledge of the controlled system. The algorithm starts with zero fuzzy rules (zero data clouds) and evolves/learns during the process control. Also the PID parameters of the controller are initialed with zeros and are adapted in online manner. According to the zero initialization of the parameters the new adaptation law is proposed in this article to solve the problems in the starting phase of the process control. Several initial scenarios were theoretically propagated and experimentally tested on the model of a heat-exchanger plant. These experiments prove that the proposed adaptation law improve the performance of the RECCo control algorithm in the starting phase.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115265282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A constructive search algorithm for combinatorial dynamic optimization problems","authors":"A. Baykasoğlu, F. Özsoydan","doi":"10.1109/EAIS.2015.7368783","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368783","url":null,"abstract":"In most of the optimization studies, the problem related data is assumed to be exactly known beforehand and remain stationary throughout whole optimization process. However, majority of real life problems and their practical applications are dynamic in their nature due to the reasons arising from unpredictable events, such as rush orders, fluctuating capacities of manufacturing constraints, changes in costs or profits. A problem, carrying one of these features is known as dynamic optimization problem (DOP) in the related literature. In DOPs the aim is not only to find the optimum of the current problem configuration, but to keep track of the moving optima. Dynamic optimization is a hot research area and a notable variety of solution methodologies are developed for DOPs in the past decade. As a contribution to the existing literature of DOPs, in the current work, the idea of using a multi-start and constructive search algorithm and thus breaking the dependency to the previously found solutions is presented. The performance tests are conducted on the generalized assignment problem, which has numerous real life applications. In regard to the obtained results, the proposed method is found promising.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117097971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modelling stock-market investors as Reinforcement Learning agents","authors":"A. Pastore, Umberto Esposito, E. Vasilaki","doi":"10.1109/EAIS.2015.7368789","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368789","url":null,"abstract":"Decision making in uncertain and risky environments is a prominent area of research. Standard economic theories fail to fully explain human behaviour, while a potentially promising alternative may lie in the direction of Reinforcement Learning (RL) theory. We analyse data for 46 players extracted from a financial market online game and test whether Reinforcement Learning (Q-Learning) could capture these players behaviour using a riskiness measure based on financial modeling. Moreover we test an earlier hypothesis that players are “naíve” (short-sighted). Our results indicate that Reinforcement Learning is a component of the decision-making process. We also find that there is a significant improvement of fitting for some of the players when using a full RL model against a reduced version (myopic), where only immediate reward is valued by the players, indicating that not all players are naíve.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129253217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Kalman filtering with a new state-space model for three-phase systems: Application to the identification of symmetrical components","authors":"A. Phan, Gilles Hermann, P. Wira","doi":"10.1109/EAIS.2015.7368807","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368807","url":null,"abstract":"In electrical energy transportation challenges, power quality issues like reducing harmonic pollution, reactive power and load unbalance, necessarily have to identify the symmetrical components of the three phases in a fast and precise way. This paper introduces a new state-space model to be used with an Extended Kalman Filter (EKF) in order to estimate in real-time the symmetrical components of distorted and time-changing power systems. The proposed model is therefore able to detect and to quantify the unbalance of general three-phase power systems. Indeed, the symmetrical components of the power system, i.e., their amplitude and phase angle values, can be deduced at each iteration from the proposed state-space model. The effectiveness of the method has been evaluated. Results and comparisons of online symmetrical components identification show the efficiency of the proposed method for disturbed and changing power systems.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116253487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A new computer control system for mental stress management using fuzzy logic","authors":"E. El-Samahy, M. Mahfouf, L. A. Torres-Salomao, J. Anzurez-Marín","doi":"10.1109/EAIS.2015.7368785","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368785","url":null,"abstract":"Mental stress is an important aspect that can, more often than not, affect the humans' performance when they attempt tasks of varying levels of complexity. Indeed, exposing the human to high levels of mental stress which he/she cannot tolerate, may affect the successful completion of the task to be accomplished. Therefore, the need for a system which can monitor the levels of the applied mental stress on human according to his/her current state is of prime importance. Hence, this work introduces a new system that can be used to automatically modify the applied mental stress according to the output of a linguistic model that employs fuzzy sets in the form of Mamdani-type fuzzy rules. This model accepts two markers which were derived based on heart rate and pupil dilation signals and predicts a performance index which is the test accuracy. Consequently, a “crisp” decision is made according to such predicted performance to alter the applied mental stress level. This system is implemented on two personal computers connected through a network; one for applying mental stress (in the form of arithmetic test), while the other is for monitoring the human mental state through the two markers and managing the mental stress levels. The results show a close match between the system's closed loop operation and the actual measurements acquired from human volunteers.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128846855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Possible use of evolving c-regression clustering for energy consumption profiles classification","authors":"D. Dovžan, I. Škrjanc","doi":"10.1109/EAIS.2015.7368792","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368792","url":null,"abstract":"In this paper an idea for classification of energy consumption profiles using an evolving c-regression method is presented. Cluster prototypes (centers) are usually defined as a mean of data around the center. The cluster center is a vector of numerical values. The method presented in this paper uses Takagi-Sugeno fuzzy models as a cluster prototype. Beside the method description also preliminary results of energy consumption profiles classification are given.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115741578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Solving Stochastic Root-Finding with adaptive d-ary search","authors":"A. Yazidi, B. Oommen","doi":"10.1109/EAIS.2015.7368782","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368782","url":null,"abstract":"The most fundamental problem encountered in the field of stochastic optimization, is the Stochastic Root Finding (SRF) problem where the task is to locate an unknown point x* for which g(x*) = 0 for a given function g that can only be observed in the presence of noise. The vast majority of the state-of-the-art solutions to the SRF problem involve the theory of stochastic approximation. The premise of the latter family of algorithms is to operate by means of so-called “small-step” processes that explore the search space in a conservative manner. Using this paradigm, the point investigated at any time instant is in the proximity of the point investigated at the previous time instant, rendering the convergence towards the optimal point, x*, to be sluggish. The unfortunate thing about such a search paradigm is that although g() contains information using which large sections of the search space can be eliminated, this information is unutilized. This paper provides a pioneering and novel scheme to discover and utilize this information. Our solution recursively shrinks the search space by, at least, a factor of 2d/3 at each epoch, where d ≥ 2 is a user-defined parameter of the algorithm. This enhances the convergence significantly. Conceptually, this is achieved through a subtle re-formulation of SRF problem in terms of a continuous-space generalization of the Stochastic Point Location (SPL) problem originally proposed by Oommen. Our scheme is based, in part, on the Continuous Point Location with Adaptive d-ary Search (CPL-AdS), originally presented. The solution to the CPL-AdS, however, is not applicable in our particular domain because of the inherent asymmetry of the SRF problem. Our solution invokes a CPL-AdS-like solution to partition the search interval into d subintervals, evaluates the location of the unknown root x* with respect to these sub-intervals using learning automata, and prunes the search space in each iteration by eliminating at least one partition. Our scheme, the CPL-AdS algorithm for SRF, denoted as SRF-AdS, is shown to converge to the unknown root x* with an arbitrary large degree of accuracy, i.e., with a probability as close to unity as desired. Unlike the classical formulation of the SPL problem proposed by Oommen et al, in our setting, the probability, p, of the “environment” suggesting an accurate response is non-constant. In fact, the latter probability depends of the point x being examined and the region that is a candidate to be pruned. The fact that p is not constant renders the analysis much more involved. The decision rules for pruning are also different from those encountered when p is constant.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"218 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123150726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving cells recognition by local database categorization in Artificial Immune System algorithm. Application to breast cancer diagnosis","authors":"Rima Daoudi, K. Djemal, A. Benyettou","doi":"10.1109/EAIS.2015.7368784","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368784","url":null,"abstract":"In this work, a hybrid classification system based local database categorization is proposed for breast cancer classification. The proposed approach aims to improve the classification rate of the Artificial Immune System (AIS) and reduce its computational time. The principle of the hybrid classifier based AIS consists in categorizing the cells sets in multiple local clusters using k-means algorithm and learning each cluster by the Radial Basis Function Neural Network. The goal of the categorization of data is to reduce the number of tests performed by each training example in AIS algorithms to select the nearest cell to be cloned which improves the cells recognition. The results obtained on the Digital Database for Screening Mammography (DDSM) show the effectiveness of the proposed classifier either in classification accuracy or computing costs compared to other AIS algorithms.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124636386","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Keynote speaker 2: Real time data mining","authors":"João Gama","doi":"10.1109/EAIS.2015.7368772","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368772","url":null,"abstract":"Nowadays, there are applications in which the data are modelled best not as persistent tables, but rather as transient data streams. In this keynote, we discuss the limitations of current machine learning and data mining algorithms. We discuss the fundamental issues in learning in dynamic environments like learning decision models that evolve over time, learning and forgetting, concept drift and change detection. Data streams are characterized by huge amounts of data that introduce new constraints in the design of learning algorithms: limited computational resources in terms of memory, processing time and CPU power. In this talk, we present some illustrative algorithms designed to taking these constrains into account. We identify the main issues and current challenges that emerge in learning from data streams, and present open research lines for further developments.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124803456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolutionary antenna design for intelligent systems networks","authors":"V. Shakhnov, L. Zinchenko, V. Verstov, B. Sorokin","doi":"10.1109/EAIS.2015.7368799","DOIUrl":"https://doi.org/10.1109/EAIS.2015.7368799","url":null,"abstract":"Communication in intelligent distributed systems plays a crucial role for their correct performance. In the paper, we discuss several approaches to communication systems design that can be used in distributed intelligent MEMS. After a brief overview of our approach practical illustration is demonstrated for design of several antennas. Conclusions are summarized on the role of penalty coefficients and fitness functions for effective evolutionary antenna design.","PeriodicalId":325875,"journal":{"name":"2015 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116626813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}