R.A. Casas

Papers

Total Citations

H-Index

About

R.A. Casas is a researcher whose work sits at the intersection of deep learning and efficient model deployment, with a primary focus on compressing convolutional neural networks (CNNs) for image classification without sacrificing accuracy. Their most cited paper, "Methodologies of Compressing a Stable Performance Convolutional Neural Networks in Image Classification" (2019, 24 citations), introduces a suite of techniques—including pruning, quantization, and knowledge distillation—that enable high-performing CNNs to operate on resource-constrained devices. This contribution is particularly valuable for real-world applications in mobile vision, edge computing, and embedded systems, where computational efficiency is critical. Casas’s work addresses a key bottleneck in AI deployment: balancing model size and speed with classification stability. By systematically evaluating compression methodologies, they have provided a practical roadmap for engineers and researchers seeking to shrink state-of-the-art networks while retaining their predictive power. Their research continues to influence the growing field of lightweight neural architectures, making deep learning more accessible and sustainable in production environments.