Author: Community

Advertisement, IEEE. IEEE Collabratec is a new, integrated online community where IEEE members, researchers, authors, and technology professionals with similar fields of interest can network and collaborate, as well as create and manage content. Featuring a suite of powerful online networking and collaboration tools, IEEE Collabratec allows you to connect according to geographic location, technical interests, or career pursuits. You can also create and share a professional identity that showcases key accomplishments and participate in groups focused around mutual interests, actively learning from and contributing to knowledgeable communities. All in one place! Learn about IEEE Collabratec at ieeecollabratec.org.

Convolutional neural networks (CNNs) are the backbones of deep learning paradigms for numerous vision tasks. Early advancements in CNN architectures are primarily driven by human expertise and by elaborate design processes. Recently, neural architecture search was proposed with the aim of automating the network design process and generating task-dependent architectures. While existing approaches have achieved competitive performance in image classification, they are not well suited to problems where the computational budget is limited for two reasons: 1) the obtained architectures are either solely optimized for classification performance, or only for one deployment scenario and 2) the search process requires vast computational resources in most approaches. To overcome these limitations, we propose an evolutionary algorithm for searching neural architectures under multiple objectives, such as classification performance and floating point operations (FLOPs). The proposed method addresses the first shortcoming by populating a set of architectures to approximate the entire Pareto frontier through genetic operations that recombine and modify architectural components progressively. Our approach improves computational efficiency by carefully down-scaling the architectures during the search as well as reinforcing the patterns commonly shared among past successful architectures through Bayesian model learning. The integration of these two main contributions allows an efficient design of architectures that are competitive and in most cases outperform both manually and automatically designed architectures on benchmark image classification datasets: CIFAR, ImageNet, and human chest X-ray. The flexibility provided from simultaneously obtaining multiple architecture choices for different compute requirements further differentiates our approach from other methods in the literature.

Provides a listing of current staff, committee members and society officers.

Provides a listing of current committee members and society officers.

In quality-diversity algorithms, the behavioral diversity metric is a key design choice that determines the quality of the evolved archives. Although behavioral diversity is traditionally obtained by describing the observed resulting behavior of robot controllers evaluated in a single environment, it is often more easily induced by introducing environmental diversity, i.e., by manipulating the environments in which the controllers are evaluated. This article proposes quality-environment-diversity (QED), an algorithm that repeatedly generates a random environment according to a probability distribution over environmental features (e.g., number of obstacles, arena size and robot sensor and actuator characteristics), evaluates the controller in that environment, and then describes the controller in terms of the features of that environment, the environment descriptor. Our study compares QED to three baseline task-specific and generic behavioral descriptors, in 5 different robot swarm benchmark tasks. For each task, the quality of the evolved archives is assessed by their capability to provide high-performing compensatory behaviors following injection of 250 unique faults to the robots of the swarm. The evolved archives achieve a median 2- to 3-fold reduction in the impact of the faults on the performance of the swarm. A qualitative analysis of evolved archives is done by visualizing the relation between diversity of compensatory behaviors, here called useful behavioral diversity, and fault recovery metrics. The resulting signatures indicate that, due to the diversity of environments inducing useful behavioral diversity, archives evolved by QED provide robot swarm controllers that are capable of recovering from high-impact faults.

To handle different types of many-objective optimization problems (MaOPs), many-objective evolutionary algorithms (MaOEAs) need to simultaneously maintain convergence and population diversity in the high-dimensional objective space. In order to balance the relationship between diversity and convergence, we introduce a Kernel matrix and probability model called determinantal point processes (DPPs). Our MaOEA with DPPs (MaOEADPPs) is presented and compared with several state-of-the-art algorithms on various types of MaOPs with different numbers of objectives. The experimental results demonstrate that MaOEADPPs is competitive.

The performance of deep neural networks is heavily dependent on its architecture and various neural architecture search strategies have been developed for automated network architecture design. Recently, evolutionary neural architecture search (EvoNAS) has received increasing attention due to the attractive global optimization capability of evolutionary algorithms. However, EvoNAS suffers from extremely high computational costs because a large number of performance evaluations are usually required in evolutionary optimization, and training deep neural networks is itself computationally very expensive. To address this issue, this article proposes a computationally efficient framework for the evolutionary search of convolutional networks based on a directed acyclic graph, in which parents are randomly sampled and trained on each mini-batch of training data. In addition, a node inheritance strategy is adopted so that the fitness of all offspring individuals can be evaluated without training them. Finally, we encode a channel attention mechanism in the search space to enhance the feature processing capability of the evolved neural networks. We evaluate the proposed algorithm on the widely used datasets, in comparison with 30 state-of-the-art peer algorithms. Our experimental results show that the proposed algorithm is not only computationally much more efficient but also highly competitive in learning performance.

The recently introduced real-valued gene-pool optimal mixing evolutionary algorthm (RV-GOMEA) has been shown to be among the state of the art for solving gray-box optimization problems where partial evaluations can be leveraged. A core strength is its ability to effectively exploit the linkage structure of a problem, which often is unknown a priori and has to be learned online. Previously published work on RV-GOMEA, however, demonstrated excellent scalability when the linkage structure is prespecified appropriately. A mutual information-based metric to learn linkage structure online, as commonly adopted in EDA’s and the original discrete version of the gene-pool optimal mixing evolutionary algorithm, did not lead to similarly excellent results, especially in a black-box setting. In this article, the strengths of RV-GOMEA are combined with a new fitness-based linkage learning approach that is inspired by differential grouping that reduces its computational overhead by an order of magnitude for problems with fewer interactions. The resulting new version of RV-GOMEA achieves scalability similar to when a predefined linkage model is used, outperforming also, for the first time, the EDA AMaLGaM upon which it is partially based in a black-box setting where partial evaluations cannot be leveraged.¹¹

This article is extended from the MSc thesis of Chantal Olieman, available at https://repository.tudelft.nl/ [24].

Anikó Ekárt and Peter R. Lewis will guest edit a special issue on Trust, Trustworthiness, and Evolvable Systems. 

See the full call for papers at https://www.springer.com/journal/10710/updates/18975534.