ブライアスコアに類似した平均絶対誤差の名前？

昨日の質問イベントの確率を推定するモデルの精度を決定することで、確率スコアリングに興味を持ちました。

\frac{1}{N} \sum_{i = 1}^{N} (p r e d i c t i o n_{i} - r e f e r e n c e_{i})^{2}

$\frac{1}{N}\sum\limits _{i=1}^{N}(prediction_i - reference_i)^2$

\frac{1}{N} \sum_{i = 1}^{N} | p r e d i c t i o n_{i} - r e f e r e n c e_{i} |

$\frac{1}{N}\sum\limits _{i=1}^{N}|prediction_i - reference_i|$ have a name, too?

— cbeleites supports Monica
ソース

Google allowed me to find this paper where something very similar is named

L_{1}

$L_1$ -calibration score. Note that this score is a bit different than yours, anyway "

L_{1}

$L_1$ score" seems the good keyword.

— Elvis

What search terms did you use? Googling I mainly learned how many different tumour scores exist (L1 meaning lymphnode involvement in that context)...

— cbeleites supports Monica

Something like "L1 score probability"... may be I've been lucky

— Elvis

Or google tries to help me and thinks I'm looking for tumours because that's what I do more often... "probability near score L1" got me to the paper below.

— cbeleites supports Monica

Answer seems to be: no, because MAE doesn't lead to a proper scoring rule.

See Loss Functions for Binary Class Probability Estimation and Classification: Structure and Applications where the MAE is discussed under "Counterexamples of proper scoring rules".

— cbeleites supports Monica
ソース

@amoeba: There you are

— cbeleites supports Monica