A learning strategy for fuzzy neural tree is presented that is based on combining the knowledge-driven and data-driven modeling paradigms. The knowledge-driven aspect of the strategy is expressing knowledge via the connection topology of a neural tree. The tree is driven by inputs associated with fuzzy logic. In this type of neural tree, the connection weights are determined in an unsupervised manner. However, the fuzzy logic related parameters are subject to data-driven identification, and they are comparatively few in number. For this reason, a low number of input-output data-pairs suffice to establish the neural representation in the new approach. This makes it suitable for representing evaluation processes of mind that have been difficult to bring into explicit form. An example of this is the evaluation of shape quality in an architectural design, and it is used to verify the effectiveness of the approach by experiment.

@en

In a previous publication by the same authors, a computational design system was described that identifies aesthetical color combinations. In that work the type of color aesthetics pursued was to be determined prior to the design representing a given design objective. Complementing the previous study, in this work, the type of color aesthetics that is most suitable for a given scene at hand is pursued taking into account the color and geometry of an existing situation. This is accomplished by bringing the parameter that characterizes the type of aesthetics into computational design, i.e. treating it as one of the components of the decision variable vector subject to identification by multi-objective evolutionary search. The parameter's influence on aesthetics is investigated theoretically, as well as by means of computer experiments. The contribution of the study to Architecture is provision of a firm base for some common architectural knowledge as to the color aesthetics of buildings. In particular light is shed on the aesthetical dependence of a building's color to the color of its environment.

@en

Studies on computer-based visual perception and aesthetical judgment for architectural design are presented. In the model, both color and the geometric aspects of human vision are jointly taken into account, quantifying the perception of an individual object, as well as a scene consisting of several objects. This is accomplished by fuzzy neural tree processing. Based on the perception model, aesthetical color compositions are identified for a scene using multi-objective evolutionary algorithm. The methodology is described together with associated computer experiments verifying the theoretical considerations. Modeling of aesthetical judgment is a significant step for
applications, where human-like visual perception and cognition are of concern. Examples of such applications are architectural design, product design, and urbanism.@en

Demonstrative results of a probabilistic constraint handling approach that is exclusively using evolutionary computation are presented. In contrast to other works involving the same probabilistic considerations, in this study local search has been omitted, in order to assess the necessity of this deterministic local search procedure in connection with the evolutionary one. The precision stems from the non-linear probabilistic distance measure that maintains stable evolutionary selection pressure towards the feasible region throughout the search, up to micro level in the range of 10-10 or beyond. The details of the theory are revealed in another paper [1]. In this paper the implementation results are presented, where the non-linear distance measure is used in the ranking of the solutions for effective tournament selection. The test problems used are selected from the existing literature. The evolutionary implementation without local search turns out to be already competitively accurate with sophisticated and accurate state-of-the-art constrained optimization algorithms. This indicates the potential for enhancement of the sophisticated algorithms, as to their precision and accuracy, by the integration of the proposed approach.@en

Comprehension of aesthetical color characteristics based on a computational model of visual perception and color cognition are presented. The computational comprehension is manifested by the machine’s capability of instantly assigning appropriate colors to the objects perceived. They form a scene with aesthetically pleasing characteristics. The present approach to computational cognition is principally the same as contrived earlier [1]. This work distinguishes itself from the earlier work through the involvement of color differences. The color difference computations are carried out based on a standard human color observer model. The color difference information is
combined with geometric perception information using the method of fuzzy neural tree based on likelihood. The study exemplifies the suitability of the computational cognition for modeling cognition phenomenon. Cognitive color perception in computational form has generic relevance to applications involving human-like aesthetical appreciation, as is the case in building architecture, for instance and other design tasks.@en

A robust probabilistic constraint handling approach in the framework of joint evolutionary-classical optimization has been presented earlier. In this work, the
theoretical foundations of the method are presented in detail. The method is known as bi-objective method, where the conventional penalty function approach is implemented. The present work highlights the dynamic variation of the commensurate penalty parameter for each objective treated as constraint. It is shown that the constraint parameters collectively define the right slope of the tangent as to the optimal front during the search. The robust and sustained convergence throughout the search up to micro level in the range of 10-10 or beyond is explained. The work here is presented as a further note in connection with the previous publication, where the subtle theoretical considerations and
their details had been omitted for the sake of detailed results of the experiments demonstrating the effective working of the approach. In contrast to the implementation-centered reporting of the previous work, this work can be considered as a description of the detailed probabilistic basis underlying the previous work. Therefore, this study is of great importance to let the researchers conveniently gain the insight into the work and its implications reported earlier.@en