Antonino Santos

Prediction and modeling of the rainfall-runoff transformation of a typical urban basin using ann and gp

This paper proposes an application of Genetic Programming (GP) and Artificial Neural Networks (ANN) in hydrology, showing how these two techniques can work together to solve a problem, namely for modeling the effect of rain on the runoff flow in a typical urban basin. The ultimate goal of this research is to design a real-time alarm system to warn of floods or subsidence in various types of urban basins. Results look promising and appear to offer some improvement for analyzing river basin systems over stochastic methods such as unitary hydrographs.

Partially interactive evolutionary artists

User fatigue is probably the most pressing problem in current Interactive Evolutionary Computation systems. To address it we propose the use of automatic seeding procedure, phenotype filters, and partial automation fitness assignment. We test this approaches in the visual arts domain. To further enhance interactive evolution applications in aesthetic domains, we propose the use of artificial art critics—systems that perform stylistic and aesthetic valuations of art—presenting experimental results.

On the development of critics in evolutionary computation artists

One of the problems in the use of evolutionary computer systems in artistic tasks is the lack of artificial models of human critics. In this paper, based on the state of the art and on our previous related work, we propose a general architecture for an artificial art critic, and a strategy for the validation of this type of system. The architecture includes two modules: the analyser, which does a pre-processing of the artwork, extracting several measurements and characteristics; and the evaluator, which, based on the output of the analyser, classifies the artwork according to a certain criteria. The validation procedure consists of several stages, ranging from author and style discrimination to the integration of critic in a dynamic environment together with humans.

Prediction and Modelling of the Flow of a Typical Urban Basin through Genetic Programming

Genetic Programming (GP) is an evolutionary method that creates computer programs that represent approximate or exact solutions to a problem. This paper proposes an application of GP in hydrology, namely for modelling the effect of rain on the run-off flow in a typical urban basin. The ultimate goal of this research is to design a real time alarm system to warn of floods or subsidence in various types of urban basin. Results look promising and appear to offer some improvement over stochastic methods for analysing river basin systems such as unitary radiographs.

Computerized measures of visual complexity.

Visual complexity influences peoples perception of, preference for, and behaviour toward many classes of objects, from artworks to web pages. The ability to predict peoples impression of the complexity of different kinds of visual stimuli holds, therefore, great potential for many domains, basic and applied. Here we use edge detection operations and several image metrics based on image compression error and Zipfs law to estimate the visual complexity of images. The experiments involved 800 images, each previously rated by thirty participants on perceived complexity. In a first set of experiments we analysed the correlation of individual features with the average human response, obtaining correlations up to rs = .771. In a second set of experiments we employed Machine Learning techniques to predict the average visual complexity score attributed by humans to each stimuli. The best configurations obtained a correlation of rs = .832. The average prediction error of the Machine Learning system over the set of all stimuli was .096 in a normalized 0 to 1 interval, showing that it is possible to predict, with high accuracy human responses. Overall, edge density and compression error were the strongest predictors of human complexity ratings.

Using complexity estimates in aesthetic image classification

In recent years, the search for computational systems that classify images based on aesthetic properties has gained momentum. Such systems have a wide range of potential applications, including image search, organization, acquisition and generation. This work explores the use of complexity estimates to predict the aesthetic merit of photographs. We use a set of image metrics and two different classifiers. Our approach classifies images gathered from a photography web site, attempting to reproduce the evaluation made by a group of users. For this purpose, we use complexity estimate metrics based on the encoding size and compression error of JPEG and fractal compression, which are applied to the original value channel and to the images resulting from applying Sobel and Canny filters to this channel. By employing these estimates, in conjunction with the average and standard deviation of the value channel, i.e., 20 features, a success rate of 74.59% was attained. Using the three most influential features yields a success rate of 71.34%, which is competitive with the best results reported in the literature, 71.44%, using the same dataset.

Technical Section: On the development of evolutionary artificial artists

The creation and the evaluation of aesthetic artifacts are tasks related to design, music and art, which are highly interesting from the computational point of view. Nowadays, Artificial Intelligence systems face the challenge of performing tasks that are typically human, highly subjective, and eventually social. The present paper introduces an architecture which is capable of evaluating aesthetic characteristics of artifacts and of creating artifacts that obey certain aesthetic properties. The development methodology and motivation, as well as the results achieved by the various components of the architecture, are described. The potential contributions of this type of systems in the context of digital art are also considered.

Automatic recurrent ANN rule extraction with genetic programming

Various rule-extraction techniques using ANNs have been used so far, most of them being applied on multi-layer ANNs, since they are more easily handled. In many cases, extraction methods focusing on different types of networks and training have been implemented, however, there are virtually no methods that view the extraction of rules from ANNs as systems which are independent from their architecture, training and internal distribution of weights, connections and activation functions. This paper proposes a rule-extraction system of ANNs regardless of their architecture (multi-layer or recurrent), using genetic programming as a rule-exploration technique.

On the Socialization of Evolutionary Art

The lack of a social context is a drawback in current Interactive Evolutionary Computation systems. In application areas where cultural characteristics are particularly important, such as visual arts and music, this problem becomes more pressing. To address this issue, we analyze variants of the traditional Interactive Evolutionary Art approach --- such as multi-user, parallel and partially interactive approaches --- and present an extension of the traditional Interactive Evolutionary Computation paradigm. This extension incorporates users and systems in a Hybrid Society model, that allows the interaction between multiple users and systems, establishing n *** m relations among them, and promotes cooperation.

Automatic Recurrent and Feed-Forward ANN Rule and Expression Extraction with Genetic Programming

Various rule-extraction techniques using ANN have been used so far, most of them being applied on multi-layer ANN, since they are more easily handled. In many cases, extraction methods focusing on different types of networks and training have been implemented. However, there are virtually no methods that view the extraction of rules from ANN as systems which are independent from their architecture, training and internal distribution of weights, connections and activation functions. This paper proposes a rule-extraction system of ANN regardless of their architecture (multi-layer or recurrent), using Genetic Programming as a rule-exploration technique.