https://scholars.lib.ntu.edu.tw/handle/123456789/607390
標題: | Efficient Optimization Methods for Extreme Similarity Learning with Nonlinear Embeddings | 作者: | Yuan B Li Y.-S Quan P Lin C.-J. CHIH-JEN LIN |
關鍵字: | neural networks;newton methods;non-convex optimization;representation learning;similarity learning;Embeddings;Learning systems;Optimization;Building blockes;Hessian-vector products;Learning similarity;Loss functions;Nonlinear embedding;Optimization algorithms;Optimization method;Similarity learning;Data mining | 公開日期: | 2021 | 起(迄)頁: | 2093-2103 | 來源出版物: | Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining | 摘要: | We study the problem of learning similarity by using nonlinear embedding models (e.g., neural networks) from all possible pairs. This problem is well-known for its difficulty of training with the extreme number of pairs. For the special case of using linear embeddings, many studies have addressed this issue of handling all pairs by considering certain loss functions and developing efficient optimization algorithms. This paper aims to extend results for general nonlinear embeddings. First, we finish detailed derivations and provide clean formulations for efficiently calculating some building blocks of optimization algorithms such as function, gradient evaluation, and Hessian-vector product. The result enables the use of many optimization methods for extreme similarity learning with nonlinear embeddings. Second, we study some optimization methods in detail. Due to the use of nonlinear embeddings, implementation issues different from linear cases are addressed. In the end, some methods are shown to be highly efficient for extreme similarity learning with nonlinear embeddings. ? 2021 ACM. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85114949554&doi=10.1145%2f3447548.3467363&partnerID=40&md5=8fd57bcb7b148b8851e39d5ec8da7a1f https://scholars.lib.ntu.edu.tw/handle/123456789/607390 |
DOI: | 10.1145/3447548.3467363 |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。