Vector Quantized-Elites: Unsupervised and Problem-Agnostic Quality-Diversity Optimization

Quality-Diversity algorithms have transformed optimization by prioritizing the discovery of diverse, high-performing solutions over a single optimal result. However, traditional Quality-Diversity methods, such as MAP-Elites, rely heavily on predefined behavioral descriptors and complete prior knowledge of the task to define the behavioral space grid, limiting their flexibility and applicability. In this work, we introduce Vector Quantized-Elites (VQ-Elites), a novel Quality-Diversity algorithm that autonomously constructs a structured behavioral space grid using unsupervised learning, eliminating the need for prior task-specific knowledge. At the core of VQ-Elites is the integration of Vector Quantized Variational Autoencoders, which enables the dynamic learning of behavioral descriptors and the generation of a structured, rather than unstructured, behavioral space grid - a significant advancement over existing unsupervised Quality-Diversity approaches. This design establishes VQ-Elites as a flexible, robust, and task-agnostic optimization framework. To further enhance the performance of unsupervised Quality-Diversity algorithms, we introduce two key components: behavioral space bounding and cooperation mechanisms, which significantly improve convergence and performance. We validate VQ-Elites on robotic arm pose-reaching and mobile robot space-covering tasks. The results demonstrate its ability to efficiently generate diverse, high-quality solutions, emphasizing its adaptability, scalability, robustness to hyperparameters, and potential to extend Quality-Diversity optimization to complex, previously inaccessible domains.

Dataset: Impact Events for Structural Health Monitoring of a Plastic Thin Plate

Nowadays, more and more datasets are published towards research and development of systems and models, enabling direct comparisons, continuous improvement of solutions, and researchers engagement with experimental, real life data. However, especially in the Structural Health Monitoring (SHM) domain, there are plenty of cases where new research projects have a unique combination of structure design and implementation, sensor selection and technological enablers that does not fit with the configuration of relevant individual studies in the literature. Thus, we share the data from our case study to the research community as we did not find any relevant repository available. More specifically, in this paper, we present a novel time-series dataset for impact detection and localization on a plastic thin-plate, towards Structural Health Monitoring applications, using ceramic piezoelectric transducers (PZTs) connected to an Internet of Things (IoT) device. The dataset was collected from an experimental procedure of low-velocity, low-energy impact events that includes at least 3 repetitions for each unique experiment, while the input measurements come from 4 PZT sensors placed at the corners of the plate. For each repetition and sensor, 5000 values are stored with 100 KHz sampling rate. The system is excited with a steel ball, and the height from which it is released varies from 10 cm to 20 cm. The dataset is available in GitHub (https://github.com/Smart-Objects/Impact-Events-Dataset).