首页|Researchers Submit Patent Application, 'Bayesian Hierarchical Modeling For Low S ignal Datasets', for Approval (USPTO 20240152789)
Researchers Submit Patent Application, 'Bayesian Hierarchical Modeling For Low S ignal Datasets', for Approval (USPTO 20240152789)
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
News editors obtained the following quote from the background information suppli ed by the inventors: “In recent years, the use of artificial intelligence, inclu ding but not limited to machine learning, deep learning, etc. (referred to colle ctively herein as artificial intelligence), has exponentially increased. Broadly described, artificial intelligence refers to a wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence. Key benefits of artificial intelligence ar e its ability to process data, find underlying patterns, and/or perform real-tim e determinations. However, despite these benefits and despite the wide-ranging n umber of potential applications, practical implementations of artificial intelli gence have been hindered by several technical problems. For example, artificial intelligence often relies on large amounts of high-quality data for training. Of ten this data is referred to as a training dataset. That is, each model is gener ally trained using a dataset that captures dataset-specific factor sensitivity f or a particular population. This type of training enables the creation of models that generate accurate predictions. However, in many instances, high-quality da ta is not available in amounts large enough for effective training. Thus, a part icular dataset may include only a small sample size or have class imbalance. Sma ll sample sizes and/or class imbalance cause reliability issues for the model sp ecification leading to biased predictions that do not generalize well in many in stances. To solve that problem, model developers, in some cases, use models gene rated from a proxy dataset. This solution is problematic because the resulting m odel does not necessarily capture factor sensitivity of the low-signal populatio n.”
Artificial IntelligenceBusinessCapit al One Services LLCCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
