ftnext / MLPaperSummary

論文のサマリーをIssueに蓄積(arXivTimesリスペクト)
0 stars 0 forks source link

[1602.04938] "Why Should I Trust You?": Explaining the Predictions of Any Classifier #11

Open ftnext opened 1 year ago

ftnext commented 1 year ago

https://scrapbox.io/nikkie-memos/%22Why_Should_I_Trust_You%3F%22:_Explaining_the_Predictions_of_Any_Classifier

まとめ

どのような分類器の予測も説明できる手法LIME (Local Interpretable Model-agnostic Explanations) を提案。

予測対象のインスタンスの周辺でサンプリングし、分類器の決定関数を使って予測。 予測対象インスタンスとサンプリングされたインスタンスの近さを重みとし、解釈可能なモデルを訓練する。 解釈可能なモデルは局所的には分類器に忠実となり、解釈可能性ゆえに説明を提供できる

ftnext commented 1 year ago

まとめスライド https://twitter.com/ftnext/status/1584513377162137600