首页>
外国专利>
DYNAMIC ARTIFICIAL INTELLIGENCE/MACHINE LEARNING MODEL UPDATE, OR RETRAIN AND UPDATE, IN DIGITAL PROCESSES AT RUNTIME
DYNAMIC ARTIFICIAL INTELLIGENCE/MACHINE LEARNING MODEL UPDATE, OR RETRAIN AND UPDATE, IN DIGITAL PROCESSES AT RUNTIME
展开▼
机译:动态人工智能/机器学习模型更新或培训和更新在运行时的数字过程中
展开▼
页面导航
摘要
著录项
相似文献
摘要
Dynamically updating, or retraining and updating an artificial intelligence (AI)/machine learning (ML) model in a digital process at runtime. AI/ML model updates or retraining and updating may not require production operations to be disrupted. The step of updating and/or retraining the AI/ML model may be included as part of the digital process. AI/ML model updates can be from internal logic (eg, from evaluation of conditions, by expressions that call AI/ML models, etc.), from external requests (eg, in a finite state machine (FSM)). may be requested from external triggers, such as file changes, database data, service calls, etc.), or both. AI/ML model updates or automation of retraining and updating may be provided, wherein the software retrains and/or updates AI/ML models after (and possibly shortly after) the AI/ML models become available. are reloaded/reinitialized/reinstantiated using
展开▼
机译:在运行时在数字过程中动态更新或再次更新和更新数字过程中的人工智能(AI)/机器学习(ML)模型。 AI / ML模型更新或再培训和更新可能不需要中断生产操作。可以包括更新和/或再培训AI / ML模型的步骤作为数字过程的一部分。 AI / ML模型更新可以来自内部逻辑(例如,通过调用AI / ML模型等的表达式评估条件,从外部请求(例如,在有限状态机(FSM)中))。可以从外部触发器请求,例如文件更改,数据库数据,服务呼叫等)或两者。可以提供AI / ML模型更新或再培训和更新的自动化,其中软件在AI / ML模型中可用之后(并且可能很快)再次检索和/或更新AI / ML型号。被重新加载/重新初始化/恢复使用
展开▼