Ablation — преднамеренное удаление признака, компонента, подсистемы или процесса из системы ИИ (AI) или машинного обучения для оценки его вклада. Ablation study сравнивает поведение системы до и после удаления, часто через повторное обучение или тестирование, чтобы определить, существенно ли удаленный элемент влияет на производительность или надежность.
The removal of a component of an AI system. An ablation study aims to determine the contribution of a component to an AI system by removing the component, and then analyzing the resultant performance of the system.
A technique for evaluating the importance of a feature or component by temporarily removing it from a model. You then retrain the model without that feature or component, and if the retrained model performs significantly worse, then the removed feature or component was likely important. For example, suppose you train a classification model on 10 features and achieve 88% precision on the test set. To check the importance of the first feature, you can retrain the model using only the nine other features. If the retrained model performs significantly worse (for instance, 55% precision), then the removed feature was probably important. Conversely, if the retrained model performs equally well, then that feature was probably not that important. Ablation can also help determine the importance of: - Larger components, such as an entire subsystem of a larger ML system - Processes or techniques, such as a data preprocessing step In both cases, you would observe how the system's performance changes (or doesn't change) after you've removed the component.