@techreport{oai:ipsj.ixsq.nii.ac.jp:00182375,
 author = {石田, 隆 and ニウー, ガン and 杉山, 将 and Takashi, Ishida and Gang, Niu and Masashi, Sugiyama},
 issue = {25},
 month = {Jun},
 note = {ラベル付きデータの収集にはコストがかかるため，実世界の分類問題を解く際の大きな障壁となる．そこで，パターンが所属しないクラスを一つ指定する補ラベルの設定を考える．補ラベルの収集は，数多くの候補の中から正解ラベルを慎重に選ぶ必要がないため，通常のラベルを収集するよりも遥かに少ない労力で済む． しかし，補ラベルは通常ラベルよりも情報が少ないため，補ラベルに適した学習方法が必要となる．本論文では，損失関数が特定の対称条件を満たすとき，分類リスクの不偏推定量が補ラベルのみから得られることを示す．そして，提案法の推定誤差の上界を理論的に求め，提案法の有用性を実験的に示す．, Collecting labeled data is costly and thus is a critical bottleneck in real-world classification tasks. To mitigate the problem, we consider a complementary label, which specifies a class that a pattern does not belong to. Collecting complementary labels would be less laborious than ordinary labels since users do not have to carefully choose the correct class from many candidate classes. However, complementary labels are less informative than ordinary labels and thus a suitable approach is needed to better learn from complementary labels. In this paper, we show that an unbiased estimator of the classification risk can be obtained only from complementary labels, if a loss function satisfies a particular symmetric condition. We theoretically prove the estimation error bounds for the proposed method, and experimentally demonstrate the usefulness of the proposed algorithms.},
 title = {補ラベルからの学習},
 year = {2017}
}