ログ損失, also known as logistic loss or クロスエントロピー損失, is a performance metric used primarily in 二値分類 problems where the predicted output is a probability value ranging between 0 and 1. It quantifies the difference between the predicted probabilities and the actual class labels (0 or 1). The objective of using log loss is to evaluate how well a classification model predicts probabilities for binary outcomes, with lower values indicating better モデルのパフォーマンス.
数学的には、ログ損失は次の式を用いて計算されます：
Log Loss = -1/N * Σ [y * log(p) + (1 – y) * log(1 – p)]
ここで：
- N の総数です。
- y 実際のラベル（0または1）です。
- p 正のクラス（1）の予測確率です。
The Log Loss value ranges from 0 to infinity, where 0 indicates perfect predictions (model outputs probabilities of either 0 or 1 correctly) and larger values indicate worse performance. A model that predicts probabilities close to the true labels will have a lower log loss, while a model that predicts probabilities far 真のラベルから外れる予測はより高いログ損失をもたらします。
Log Loss is particularly useful in scenarios where the output is not just a hard classification but a probability, making it suitable for applications such as ロジスティック回帰, neural networks, and other probabilistic classifiers. It is also widely 機械学習で使用される Kaggleなどのコンペティションでモデルの性能を評価するために。