Evolving intelligent system

The term Evolving Intelligent Systems (EIS) was coined in 2005 by Angelov and Kasabov,[1][2] to define the new approach which focuses on learning, developing soft computing models that have both their parameters but also their structure adapting on-line. The first papers in this direction can be traced back to the end of 20th century,[3][4] but these were closer to evolutionary algorithms. Later, in the very beginning of the 21st century both Angelov and Kasabov came independently to evolving rule-based,[5] and evolving connectionist systems [6] concepts respectively. This was followed later by a large number of research works by various authors,[7] etc. who meet annually at IEEE sponsored conferences on this topic.

EIS are usually associated with, streaming data and on-line (often real-time) modes of operation. They can be seen as adaptive intelligent systems with low-computational complexity. EIS assumes on-line adaptation of system structure in addition to the parameter adaptation which is usually associated with the term "incremental" from Incremental heuristic search. Due to the implementation of a wide variety of adaptive, evolving and dynamic methodologies, they represent an important cornerstone within the field of data-driven learning in non-stationary environments [8]

An important sub-area of EIS is represented by Evolving Fuzzy Systems (EFS) (a comprehensive survey including real-world applications can be found in [7]), which rely on fuzzy systems architecture and incrementally update, evolve and prune fuzzy sets and fuzzy rules on demand and on-the-fly. One of the major strengths of EFS, compared to other forms of evolving system models, is that they are able to support some sort of interpretability and understandability for experts and users.[9] This opens possibilities for enriched human-machine interaction's scenarios, where the users may "communicate" with an on-line evolving system in form of knowledge exchange (active learning (machine learning) and teaching). This concept is currently motivated and discussed in the evolving systems community under the term Human-Inspired Evolving Machines and respected as "one future" generation of "EIS".[10]

Evolving versus Evolutionary

The word evolve has two definitions in the Oxford dictionary first is "to develop gradually" as a verb, which is the key function of Adaptive and Evolving Systems and the second that is "to develop over successive generations as a result of natural selection" with reference to an organism or biological feature, which is the key idea of the Evolutionary Computation.

Often evolving is used in relation to evolutionary computation. EIS consider a gradual development of the underlying system structure but do not deal with such phenomena specific for the Evolutionary algorithms or genetic algorithms as chromosomes crossover, mutation, selection and reproduction, parents and off-springs.

References

  1. ↑ P. Angelov and N. Kasabov (2005), Evolving Computational Intelligence Systems, Proc 1st International Workshop on Genetic Fuzzy Systems, Granada, Spain, pp 76-82.
  2. ↑ P. Angelov and N. Kasabov (2006), Evolving Intelligent Systems, eIS, IEEE SMC eNewsLetter, June 2006, pp.1-13.
  3. ↑ N. Kasabov (1998), Evolving Fuzzy Neural Networks—Algorithms, Applications and Biological Motivation, Proc. of Iizuka'98, Iizuka, Japan, Oct.1998, World Sci., 271– 274 (1998).
  4. ↑ P. P. Angelov (1999), Evolving Fuzzy Rule-Based Models, Proc. 8th IFSA World Congress, Taiwan, vol.1, pp.19-23.
  5. ↑ P. Angelov, R. Buswell (2001), Evolving Rule-based Models: A Tool for Intelligent Adaptation, 9th IFSA World Congress, Vancouver, BC, Canada, 25–28 July 2001, pp.1062-1067.
  6. ↑ N. Kasabov, Q. Song (2002), DENFIS: Dynamic Evolving Neural-Fuzzy Inference System and Its Application for Time-Series Prediction, IEEE Trans. on Fuzzy Systems, vol. 10, pp. 144-154.
  7. 1 2 E. Lughofer (2011), Evolving Fuzzy Systems: Methodologies, Advanced Concepts and Applications. Studies in Fuzzy and Soft Computing, Springer.
  8. ↑ M. Sayed-Mouchaweh, E. Lughofer (2012), Learning in Non-Stationary Environments: Methods and Applications. Springer, New York, 2012.
  9. ↑ E. Lughofer (2013), On-line Assurance of Interpretability in Evolving Fuzzy Systems – Achievements, New concepts and Open Issues. Information Sciences, vol. 251, pp. 22-46, 2013.
  10. ↑ E. Lughofer (2011), Human Inspired Evolving Machines - The Next Generation of Evolving Intelligent Systems?. IEEE Newsletter, vol. 36, 2011.
This article is issued from Wikipedia - version of the 6/27/2015. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.