Includes bibliographical references )p. 309-325( and index
Masashi Sugiyama, Taiji Suzuki, Takafumi Kanamori
Part I. Density-Ratio Approach to Machine Learning: 1. Introduction -- Part II. Methods of Density-Ratio Estimation: 2. Density estimation; 3. Moment matching; 4. Probabilistic classification; 5. Density fitting; 6. Density-ratio fitting; 7. Unified framework; 8. Direct density-ratio estimation with dimensionality reduction -- Part III. Applications of Density Ratios in Machine Learning: 9. Importance sampling; 01. Distribution comparison; 11. Mutual information estimation; 21. Conditional probability estimation -- Part IV. Theoretical Analysis of Density-Ratio Estimation: 31. Parametric convergence analysis; 41. Non-parametric convergence analysis; 51. Parametric two-sample test; 61. Non-parametric numerical stability analysis -- Part V. Conclusions: 71. Conclusions and future directions