首页|'Machine Learning Based Matching Of Entities Using Multiple Queues' in Patent Ap plication Approval Process (USPTO 20240403113)
'Machine Learning Based Matching Of Entities Using Multiple Queues' in Patent Ap plication Approval Process (USPTO 20240403113)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
The following quote was obtained by the news editors from the background informa tion supplied bythe inventors: “Task scheduling systems manage distribution of tasks or jobs across multiple workers. Taskscheduling techniques often used inc lude first come first served strategy, round robin scheduling strategy,greedy s cheduling strategies, and so on. Conventional techniques often have drawbacks ma king themunsuitable for certain applications. For example, greedy scheduling st rategies make locally optimal choiceat each step and may find suboptimal soluti ons. First come first served task scheduling strategies aresimple to implement but may not perform well if the system aims to optimize the result based on doma inspecific criteria. In general, conventional scheduling techniques provide sub optimal results for applicationsthat optimize domain specific criteria and mana ge multiple task queues. Such task scheduling strategies have impact on the perf ormance on the overall system, thereby causing the system to fail to achieve anoverall optimal goal.”
NETL
NSTL
