Local Rademacher Complexity for Multi-Label Learning
From MaRDI portal
Abstract: We analyze the local Rademacher complexity of empirical risk minimization (ERM)-based multi-label learning algorithms, and in doing so propose a new algorithm for multi-label learning. Rather than using the trace norm to regularize the multi-label predictor, we instead minimize the tail sum of the singular values of the predictor in multi-label learning. Benefiting from the use of the local Rademacher complexity, our algorithm, therefore, has a sharper generalization error bound and a faster convergence rate. Compared to methods that minimize over all singular values, concentrating on the tail singular values results in better recovery of the low-rank structure of the multi-label predictor, which plays an import role in exploiting label correlations. We propose a new conditional singular value thresholding algorithm to solve the resulting objective function. Empirical studies on real-world datasets validate our theoretical results and demonstrate the effectiveness of the proposed algorithm.
Cited in
(8)- Kernel negative \(\varepsilon\) dragging linear regression for pattern classification
- Label distribution learning by regularized sample self-representation
- Improving localized multiple kernel learning via radius-margin bound
- Joint ranking SVM and binary relevance with robust low-rank learning for multi-label classification
- An adaptive fixed-point proximity algorithm for solving total variation denoising models
- Optimal ranking in multi-label classification using local precision rates
- Multi-kernel learning for multi-label classification with local Rademacher complexity
- A constrained least squares regression model
This page was built for publication: Local Rademacher Complexity for Multi-Label Learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4616316)
