ABSTRACT
Since their inception, active interactive genetic algorithms have successfully combat user evaluation fatigue induced by repetitive evaluation. Their success originates on building models of the user preferences based on partial-order graphs to create a numeric synthetic fitness. Active interactive genetic algorithms can easily reduce up to seven times the number of evaluations required from the user by optimizing such a synthetic fitness. However, despite basic understanding of the underlying mechanisms there is still a lack of principled understanding of what properties make a partial ordering graph a successful model of user preferences. Also, there has been little research conducted about how to integrate together the contributions of different users to successfully capitalize on parallelized evaluation schemes. This paper addresses both issues describing: (1) what properties make a partial-order graph successful and accurate, and (2) how partial-order graphs obtained from different users can be merged meaningfully.
- Alías, F., Llorà, X., Formiga, L., Sastry, K., and D.E., G. Efficient interactive weight tuning for TTS synthesis: reducing user fatigue by improving user consistency. In Proceedings of ICASSP (Toulouse, France, May 2006), vol. I, pp. 865--868.Google ScholarCross Ref
- Alm, C. O., and Llorà, X. Evolving emotional prosody. In Proceedings of the Ninth International Conference on Spoken Language Processing (INTERSPEECH 2006) (2006), p. paper 1741.Google Scholar
- Caldwell, C., and Johnston, V. S. Tracking a criminal suspect through face-space with a genetic algorithm. In Proceedings of the Fourth International Conference on Genetic Algorithms (1991), Morgan Kaufmann, pp. 416--421.Google Scholar
- Clark, J., and Holton, D. A. A First Look at Graph Theory. World Scientific, 1991.Google ScholarCross Ref
- Coello-Coello, C. A. An updated survey of GA-Based Multiobjective Optimization Techniques. Technical report lania-rd-09-08, Laboratorio Nacional de Informática Avanzada (LANIA), Xalapa, Veracruz, México, December, 1998.Google Scholar
- Cristianini, N., and Shawe-Taylor, J. An Introduction to Support Vector Machines. Cambridge Press, 2000. Google ScholarDigital Library
- Dawkins, R. The blind watchmaker. New York: W. W. Norton, 1986.Google Scholar
- Deb, K., Agrawal, S., Pratab, A., and Meyarivan, T. A Fast Elitist Non-Dominated Sorting Genetic Algorithm for Multi-Objective Optimization: NSGA-II. KanGAL report 200001, Indian Institute of Technology, 2000.Google Scholar
- Formiga, L., and Alías, F. Extracting user Preferences by GTM for aiGA Weight Tuning in Unit Selection Text-to-Speech Synthesis. Lecture Notes in Computer Science, 4507 (June 2007), 654--651. Google ScholarDigital Library
- Goldberg, D. E. Genetic algorithms in search, optimization, and machine learning. Adison-Wesley, 1989. Google ScholarDigital Library
- Goldberg, D. E. The design of innovation: Lessons from and for competent genetic algorithms. Kluwe Academic Publisher, 2002. Google ScholarDigital Library
- Goldberg, D. E., Korb, B., and Deb, K. Messy genetic algorithms: Motivation, analysis, and first results. Complex Systems 3, 5 (1989), 493--530.Google Scholar
- Harary, F. Graph Theory. Addison-Wesley Publishing Company, 1971.Google Scholar
- Harik, G., Lobo, F., and Goldberg, D. E. The compact genetic algorithm. Proceedings of the IEEE International Conference on Evolutionary Computation (1998), 523--528. (Also IlliGAL Report No. 97006).Google ScholarCross Ref
- Llorà, X., Goldberg, D. E., Ohsawa, Y., Matsumura, N., Washida, Y., Tamura, H., Yoshikawa, M., Welge, M., Auvil, L., Searsmith, D., Ohnishi, K., and Chao, C.-J. Innovation and Creativity Support via Chance Discovery, Genetic Algorithms, and Data Mining. Journal of New Mathematics and Natural Computation 2, 1 (2006), 85--100.Google Scholar
- Llorà, X., Sastry, K., Alıas, F., Goldberg, D. E., and Welge, M. Analyzing active interactive genetic algorithms using visual analytics. In GECCO '06: Proceedings of the 8th annual conference on Genetic and evolutionary computation (New York, NY, USA, 2006), ACM, pp. 1417--1418. Google ScholarDigital Library
- Llorà, X., Sastry, K., Goldberg, D. E., Gupta, A., and Lakshmi, L. Combating user fatigue in iGAs: Partial ordering, support vector machines, and synthetic fitness. In Proceedings of Genetic and Evolutionary Computation Conference 2005 (GECCO'05) (Washington DC, USA, June 2005), ACM, pp. 1363--1370. Google ScholarDigital Library
- Mitchell, T. M. Machine Learning. McGraw Hill, 1997. Google ScholarDigital Library
- Sastry, K. Evaluation-relaxation schemes for genetic and evolutionary algorithms. Master's thesis, University of Illinois at Urbana-Champaign, General Engineering Department, Urbana, IL, 2001. (Also IlliGAL Report No. 2002004).Google Scholar
- Shawe-Taylor, J., and Cristianini, N. Kernel Methods for Patter Analysis. Cambridge Press, 2004. Google ScholarDigital Library
- Takagi, H. Interactive evolutionary computation: Fusion of the capabilities of ec optimization and human evaluation. In Proceedings of the IEEE (September 2001), vol. 89, pp. 1275--1296.Google ScholarCross Ref
- Witten, I. H., and Frank, E. Data Mining: practical machine learning tools and techniques with java implementations. Morgan Kaufmann, 2000. Google ScholarDigital Library
Index Terms
- Graph-theoretic measure for active iGAs: interaction sizing and parallel evaluation ensemble
Recommendations
On the ensemble of optimal dominating and locating-dominating codes in a graph
Let G be a simple, undirected graph with vertex set V. For every v V , we denote by N ( v ) the set of neighbours of v, and let N v ] = N ( v ) { v } . A set C V is said to be a dominating code in G if the sets N v ] C , v V , are all nonempty. A set C ...
A Note on the Inducibility of $$4$$4-Vertex Graphs
There is much recent interest in understanding the density at which constant size graphs can appear in a very large graph. Specifically, the inducibility of a graph $$H$$H is its extremal density, as an induced subgraph of $$G$$G, where $$|G| \...
Smaller Universal Targets for Homomorphisms of Edge-Colored Graphs
Computing and CombinatoricsAbstractFor a graph G, the density of G, denoted D(G), is the maximum ratio of the number of edges to the number of vertices ranging over all subgraphs of G. For a class of graphs, the value is the supremum of densities of graphs in . A k-edge-colored ...
Comments