Generalized Learning Vector Quantization for Classification in Randomized Neural Networks and Hyperdimensional Computing
This work addresses efficiency challenges for machine learning on edge devices, offering an incremental improvement over existing RVFL networks.
The paper tackles the problem of resource-constrained machine learning on edge devices by proposing a modified Random Vector Functional Link (RVFL) network that replaces the least-squares classifier with a Generalized Learning Vector Quantization (GLVQ) classifier, achieving state-of-the-art accuracy on UCI datasets and reducing computational costs to 21% of the original.
Machine learning algorithms deployed on edge devices must meet certain resource constraints and efficiency requirements. Random Vector Functional Link (RVFL) networks are favored for such applications due to their simple design and training efficiency. We propose a modified RVFL network that avoids computationally expensive matrix operations during training, thus expanding the network's range of potential applications. Our modification replaces the least-squares classifier with the Generalized Learning Vector Quantization (GLVQ) classifier, which only employs simple vector and distance calculations. The GLVQ classifier can also be considered an improvement upon certain classification algorithms popularly used in the area of Hyperdimensional Computing. The proposed approach achieved state-of-the-art accuracy on a collection of datasets from the UCI Machine Learning Repository - higher than previously proposed RVFL networks. We further demonstrate that our approach still achieves high accuracy while severely limited in training iterations (using on average only 21% of the least-squares classifier computational costs).