Appl. Syst. Innov., Volume 7, Issue 4 (August 2024) – 3 articles

In recent years, model compression techniques have gained significant attention as a means to reduce the computational and memory requirements of deep neural networks. Knowledge distillation and pruning are two prominent approaches in this domain, each offering unique advantages in achieving model efficiency. This paper investigates the combined effects of knowledge distillation and two pruning strategies, weight pruning and channel pruning, on enhancing compression efficiency and model performance. The study introduces a metric called “Performance Efficiency” to evaluate the impact of these pruning strategies on model compression and performance. Our research is conducted on the popular datasets CIFAR-10 and CIFAR-100. We compared diverse model architectures, including ResNet, DenseNet, EfficientNet, and MobileNet. The results emphasize the efficacy of both weight and channel pruning in achieving model compression. However, a significant distinction emerges, with weight pruning showing superior performance across all four architecture types. We realized that the weight pruning method better adapts to knowledge distillation than channel pruning. Pruned models show a significant reduction in parameters without a significant reduction in accuracy. Full article

The urgent need for research and study with nondestructive and noninvasive methods and the preservation of cultural heritage led to the development and application of methodologies for the multi-level digitization of cultural elements. Photogrammetry and three-dimensional scanning offer photorealistic and accurate digital representations, while X-rays and computed tomography reveal properties and characteristics of the internal and invisible structure of objects. However, the investigation of and access to these datasets are, in several cases, limited due to the increased computing resources and the special knowledge required for their processing and analysis. The evolution of immersive technologies and the creative industry of video games offers unique user experiences. Game engines are the ideal platform to host the development of easy-to-use applications that combine heterogeneous data while simultaneously integrating immersive and emerging technologies. This article seeks to shed light on how heterogeneous digital representations of 3D imaging and tomography can be harmoniously combined in a virtual space and, through simple interactions, provide holistic knowledge and enhanced experience to end users. This research builds on previous experience concerning the virtual museum for the Antikythera Mechanism and describes a conceptual framework for the design and development of an affordable and easy-to-use display tool for combined representations of heterogeneous datasets in the virtual space. Our solution was validated by 62 users who participated in tests and evaluations. The results show that the proposed methodology met its objectives. Apart from cultural heritage, the specific methodology could be easily extended and adapted for training purposes in a wide field of application, such as in education, health, engineering, industry, and more. Full article

Data-centric artificial intelligence (DC-AI) is a modern paradigm that gives more priority to data quality enhancement, rather than only optimizing the complex codes of AI models. The DC-AI paradigm is expected to substantially advance the status of AI research and developments, which has been solely based on model-centric AI (MC-AI) over the past 30 years. Until present, there exists very little knowledge about DC-AI, and its significance in terms of solving real-world problems remains unexplored in the recent literature. In this technical note, we present the core aspects of DC-AI and MC-AI and discuss their interplay when used to solve some real-world problems. We discuss the potential scenarios/situations that require the integration of DC-AI with MC-AI to solve challenging problems in AI. We performed a case study on a real-world dataset to corroborate the potential of DC-AI in realistic scenarios and to prove its significance over MC-AI when either data are limited or their quality is poor. Afterward, we comprehensively discuss the challenges that currently hinder the realization of DC-AI, and we list promising avenues for future research and development concerning DC-AI. Lastly, we discuss the next-generation computing for DC-AI that can foster DC-AI-related developments and can help transition DC-AI from theory to practice. Our detailed analysis can guide AI practitioners toward exploring the undisclosed potential of DC-AI in the current AI-driven era. Full article