講演抄録/キーワード |
講演名 |
2011-11-09 15:45
Relative Density-Ratio Estimation for Robust Distribution Comparison ○Makoto Yamada(Tokyo Inst. of Tech.)・Taiji Suzuki(Univ. of Tokyo)・Takafumi Kanamori(Nagoya Univ.)・Hirotaka Hachiya・Masashi Sugiyama(Tokyo Inst. of Tech.) IBISML2011-46 |
抄録 |
(和) |
(まだ登録されていません) |
(英) |
Divergence estimators based on direct approximation of density-ratios
without going through separate approximation of numerator and denominator densities
have been successfully applied to machine learning tasks
that involve distribution comparison
such as outlier detection, transfer learning, and two-sample homogeneity test.
However, since density-ratio functions often possess high fluctuation,
divergence estimation is still a challenging task in practice.
In this paper, we propose to use \emph{relative divergences}
for distribution comparison,
which involves approximation of \emph{relative density-ratios}.
Since relative density-ratios are always smoother than corresponding ordinary density-ratios,
our proposed method is favorable in terms of the non-parametric convergence speed.
Furthermore, we show that the proposed divergence estimator has asymptotic variance
\emph{independent} of the model complexity under a parametric setup,
implying that the proposed estimator hardly overfits
even with complex models.
Through experiments, we demonstrate the usefulness of the proposed approach. |
キーワード |
(和) |
/ / / / / / / |
(英) |
Relative Density-Ratio / Outlier Detection / Two-Sample Test / Transfer Learning / / / / |
文献情報 |
信学技報, vol. 111, no. 275, IBISML2011-46, pp. 25-32, 2011年11月. |
資料番号 |
IBISML2011-46 |
発行日 |
2011-11-02 (IBISML) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
IBISML2011-46 |
|