[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["缺少我需要的資訊","missingTheInformationINeed","thumb-down"],["過於複雜/步驟過多","tooComplicatedTooManySteps","thumb-down"],["過時","outOfDate","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["示例/程式碼問題","samplesCodeIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間:2024-08-13 (世界標準時間)。"],[[["Equality of opportunity in machine learning focuses on ensuring that qualified individuals have an equal chance of being accepted, regardless of their demographic group."],["It's achieved when the acceptance rates for qualified individuals are the same across different demographic groups, as illustrated by the example with a 40% acceptance rate for qualified candidates in both the majority and minority groups."],["While it promotes fairness in specific scenarios, equality of opportunity has limitations, such as its dependence on a clear preferred label and potential challenges in situations lacking demographic data."],["Unlike demographic parity which focuses on overall acceptance rates, equality of opportunity concentrates on the acceptance rates within the qualified subset of each group."],["It's possible for a model to satisfy both demographic parity and equality of opportunity simultaneously, under specific conditions where positive prediction rates and true positive rates are balanced across groups."]]],[]]