Advancements in Prototype-Based Machine Learning Models

Current machine learning models compromise between accuracy and being able to explain their output. In this video, NEC Senior Research Scientist, Dr. Sascha Saralajew, highlights how NEC is reducing this trade-off by developing interpretable, fast-to-train and robust prototype-based machine learning models. These models are inspired by NEC’s pioneering contributions to machine learning and explainable AI (XAI) and human-centric AI – the Generalized Learning Vector Quantization (GLVQ) algorithm. GLVQ maintains the interpretability of basic prototype-based models, while being more powerful and achieving a higher accuracy when compared to third-party algorithms for prototype-based models. To learn more about NEC’s XAI work, visit https://neclab.eu/research-areas/data....

#NECLaboratoriesEurope #machinelearning #explainableAI #prototypebasedlearning #EUAIAct