Replenishment of AI/ML models through Action Center, artificial intelligence (AI)/machine learning (ML) model retraining hardware control, and AI/ML model configuration management

摘要

It is disclosed to supplement artificial intelligence (AI)/machine learning (ML) models, provide AI/ML model retraining hardware control, and manage AI/ML model settings through the Action Center. AI/ML models may be deployed on a hosting infrastructure where AI/ML models may be invoked by robotic process automation (RPA) robots. When the performance of the AI/ML model falls below a threshold, the results and other data from the AI/ML model predictions are sent to an action center, where humans can use the appropriate application to review the data and approve predictions, or when predictions go wrong. provide correction. These action center-approved results are then sent to the RPA robot to be used instead of predictions from the AI/ML model.