arXiv subjects

Penousal Machado

Publications and source records attributed to Penousal Machado.

43 records · Page 3Linked to original sources

arXiv2019

Fast-DENSER++: Evolving Fully-Trained Deep Artificial Neural Networks

This paper proposes a new extension to Deep Evolutionary Network Structured Evolution (DENSER), called Fast-DENSER++ (F-DENSER++). The vast majority of NeuroEvolution methods that optimise Deep Artificial Neural Networks (DANNs) only evaluate the candidate solutions for a fixed amount of epochs; this makes it difficult to effectively assess the learning strategy, and requires the best generated network to be further trained after evolution. F-DENSER++ enables the training time of the candidate solutions to grow continuously as necessary, i.e., in the initial generations the candidate solutions are trained for shorter times, and as generations proceed it is expected that longer training cycles enable better performances. Consequently, the models discovered by F-DENSER++ are fully-trained DANNs, and are ready for deployment after evolution, without the need for further training. The results demonstrate the ability of F-DENSER++ to effectively generate fully-trained DANNs; by the end of evolution, whilst the average performance of the models generated by F-DENSER++ is of 88.73%, the performance of the models generated by the previous version of DENSER (Fast-DENSER) is 86.91% (statistically significant), which increases to 87.76% when allowed to train for longer.

cs.NE↗

arXiv2019

A Pig, an Angel and a Cactus Walk Into a Blender: A Descriptive Approach to Visual Blending

A descriptive approach for automatic generation of visual blends is presented. The implemented system, the Blender, is composed of two components: the Mapper and the Visual Blender. The approach uses structured visual representations along with sets of visual relations which describe how the elements (in which the visual representation can be decomposed) relate among each other. Our system is a hybrid blender, as the blending process starts at the Mapper (conceptual level) and ends at the Visual Blender (visual representation level). The experimental results show that the Blender is able to create analogies from input mental spaces and produce well-composed blends, which follow the rules imposed by its base-analogy and its relations. The resulting blends are visually interesting and some can be considered as unexpected.

cs.AI↗

arXiv2018

Using Computer Vision Techniques for Moving Poster Design

Graphic Design encompasses a wide range of activities from the design of traditional print media (e.g., books and posters) to site-specific (e.g., signage systems) and electronic media (e.g., interfaces). Its practice always explores the new possibilities of information and communication technologies. Therefore, interactivity and participation have become key features in the design process. Even in traditional print media, graphic designers are trying to enhance user experience and exploring new interaction models. Moving posters are an example of this. This type of posters combine the specific features of motion and print worlds in order to produce attractive forms of communication that explore and exploit the potential of digital screens. In our opinion, the next step towards the integration of moving posters with the surroundings, where they operate, is incorporating data from the environment, which also enables the seamless participation of the audience. As such, the adoption of computer vision techniques for moving poster design becomes a natural approach. Following this line of thought, we present a system wherein computer vision techniques are used to shape a moving poster. Although it is still a work in progress, the system is already able to sense the surrounding physical environment and translate the collected data into graphical information. The data is gathered from the environment in two ways: (1) directly using motion tracking; and (2) indirectly via contextual ambient data. In this sense, each user interaction with the system results in a different experience and in a unique poster design.

cs.HC↗

arXiv2018

Evotype: Towards the Evolution of Type Stencils

Typefaces are an essential resource employed by graphic designers. The increasing demand for innovative type design work increases the need for good technological means to assist the designer in the creation of a typeface. We present an evolutionary computation approach for the generation of type stencils to draw coherent glyphs for different characters. The proposed system employs a Genetic Algorithm to evolve populations of type stencils. The evaluation of each candidate stencil uses a hill climbing algorithm to search the best configurations to draw the target glyphs. We study the interplay between legibility, coherence and expressiveness, and show how our framework can be used in practice.

cs.NE↗

arXiv2018

DENSER: Deep Evolutionary Network Structured Representation

Deep Evolutionary Network Structured Representation (DENSER) is a novel approach to automatically design Artificial Neural Networks (ANNs) using Evolutionary Computation. The algorithm not only searches for the best network topology (e.g., number of layers, type of layers), but also tunes hyper-parameters, such as, learning parameters or data augmentation parameters. The automatic design is achieved using a representation with two distinct levels, where the outer level encodes the general structure of the network, i.e., the sequence of layers, and the inner level encodes the parameters associated with each layer. The allowed layers and range of the hyper-parameters values are defined by means of a human-readable Context-Free Grammar. DENSER was used to evolve ANNs for CIFAR-10, obtaining an average test accuracy of 94.13%. The networks evolved for the CIFA--10 are tested on the MNIST, Fashion-MNIST, and CIFAR-100; the results are highly competitive, and on the CIFAR-100 we report a test accuracy of 78.75%. To the best of our knowledge, our CIFAR-100 results are the highest performing models generated by methods that aim at the automatic design of Convolutional Neural Networks (CNNs), and are amongst the best for manually designed and fine-tuned CNNs.

cs.NE↗

arXiv2017

Generation of concept-representative symbols

The visual representation of concepts or ideas through the use of simple shapes has always been explored in the history of Humanity, and it is believed to be the origin of writing. We focus on computational generation of visual symbols to represent concepts. We aim to develop a system that uses background knowledge about the world to find connections among concepts, with the goal of generating symbols for a given concept. We are also interested in exploring the system as an approach to visual dissociation and visual conceptual blending. This has a great potential in the area of Graphic Design as a tool to both stimulate creativity and aid in brainstorming in projects such as logo, pictogram or signage design.

cs.GR↗

arXiv2017

Towards the Evolution of Multi-Layered Neural Networks: A Dynamic Structured Grammatical Evolution Approach

Current grammar-based NeuroEvolution approaches have several shortcomings. On the one hand, they do not allow the generation of Artificial Neural Networks (ANNs composed of more than one hidden-layer. On the other, there is no way to evolve networks with more than one output neuron. To properly evolve ANNs with more than one hidden-layer and multiple output nodes there is the need to know the number of neurons available in previous layers. In this paper we introduce Dynamic Structured Grammatical Evolution (DSGE): a new genotypic representation that overcomes the aforementioned limitations. By enabling the creation of dynamic rules that specify the connection possibilities of each neuron, the methodology enables the evolution of multi-layered ANNs with more than one output neuron. Results in different classification problems show that DSGE evolves effective single and multi-layered ANNs, with a varying number of output neurons.

cs.NE↗