论文信息 - Supplemental Materials of ``Learning with Feature and Distribution Evolvable Streams''

Supplemental Materials of ``Learning with Feature and Distribution Evolvable Streams''

A.1. Rademacher Complexity In this work, we use the Rademacher complexity (Bartlett & Mendelson, 2002) in proving generalization error bounds. To simplify the presentation, we first introduce some notations. Let S = {(x1, y1), . . . , (xm, ym)} be a sample of m points drawn independently and identically distributed according to the distribution D, and then the risk and empirical risk of hypothesis g are defined as

Zhi-Hua Zhou | Yuan Jiang | Zhen-Yu Zhang | Peng Zhao

[1] M. Talagrand,et al. Probability in Banach Spaces: Isoperimetry and Processes , 1991 .

[2] Peter L. Bartlett,et al. Rademacher and Gaussian Complexities: Risk Bounds and Structural Results , 2003, J. Mach. Learn. Res..

[3] André Elisseeff,et al. Stability and Generalization , 2002, J. Mach. Learn. Res..

[4] Shai Ben-David,et al. Detecting Change in Data Streams , 2004, VLDB.

[5] Koby Crammer,et al. A theory of learning from different domains , 2010, Machine Learning.

[6] Massih-Reza Amini,et al. Learning from Multiple Partially Observed Views - an Application to Multilingual Text Categorization , 2009, NIPS.

[7] M. Ng,et al. Co-transfer learning via joint transition probability graph based method , 2012, CDKD '12.

[8] Mehryar Mohri,et al. New Analysis and Algorithm for Learning with Drifting Distributions , 2012, ALT.

[9] Ameet Talwalkar,et al. Foundations of Machine Learning , 2012, Adaptive computation and machine learning.

[10] Anton van den Hengel,et al. Image-Based Recommendations on Styles and Substitutes , 2015, SIGIR.

[11] Mehryar Mohri,et al. Adaptation Based on Generalized Discrepancy , 2019, J. Mach. Learn. Res..

[12] Zhi-Hua Zhou,et al. Learning With Feature Evolvable Streams , 2017, IEEE Transactions on Knowledge and Data Engineering.