首页|Recent Findings in Artificial Intelligence Described by Researchers from Univers ity of Colorado Denver (Towards Explainable Artificial Intelligence Through Expe rt-augmented Supervised Feature Selection)
Recent Findings in Artificial Intelligence Described by Researchers from Univers ity of Colorado Denver (Towards Explainable Artificial Intelligence Through Expe rt-augmented Supervised Feature Selection)
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
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-Researchers detail new data in Artificial Intelli gence. According to news reporting from Denver, Colorado, by NewsRx journalists, research stated, "This paper presents a comprehensive framework for expert-augm ented supervised feature selection, addressing pre-processing, in-processing, an d postprocessing aspects of Explainable Artificial Intelligence (XAI). As part of pre-processing XAI, we introduce the Probabilistic Solution Generator through the Information Fusion (PSGIF) algorithm, leveraging ensemble techniques to enh ance the exploration and exploitation capabilities of a Genetic Algorithm (GA)." The news correspondents obtained a quote from the research from the University o f Colorado Denver, "Balancing explainability and prediction accuracy, we formula te two multi -objective optimization models that empower expert(s) to specify a maximum acceptable sacrifice percentage. This approach enhances explainability b y reducing the number of selected features and prioritizing those considered mor e relevant from the domain expert ' s perspective. This contribution aligns with in-processing XAI, incorporating expert opinions into the feature selection pro cess as a multi -objective problem. Traditional feature selection techniques lac k the capability to efficiently search the solution space considering our explai nability-focused objective function. To overcome this, we leverage the Genetic A lgorithm (GA), a powerful metaheuristic algorithm, optimizing its parameters thr ough Bayesian optimization. For post-processing XAI, we present the Posterior En semble Algorithm (PEA), estimating the predictive power of features. PEA enables a nuanced comparison between objective and subjective importance, identifying f eatures as underrated, overrated, or appropriately rated. We evaluate the perfor mance of our proposed GAs on 16 publicly available datasets, focusing on predict ion accuracy in a single objective setting. Moreover, we test our multi -objecti ve model on a classification dataset to show the applicability and effectiveness of our framework."
DenverColoradoUnited StatesNorth a nd Central AmericaAlgorithmsArtificial IntelligenceEmerging TechnologiesGenetic AlgorithmsMachine LearningUniversity of Colorado Denver.
NETL
NSTL
