Continuous Optimization for Data Science
暫譯: 數據科學的持續優化

Haviv, Moshe

Continuous Optimization for Data Science
  • 出版商: World Scientific Pub
  • 出版日期: 2025-08-15
  • 售價: $4,250
  • 貴賓價: 9.5 折 $4,038
  • 語言: 英文
  • 頁數: 300
  • 裝訂: Hardcover - also called cloth, retail trade, or trade
  • ISBN: 9811299196
  • ISBN-13: 9789811299193
  • 相關分類: Data Science

    • 尚未上市，無法訂購

相關主題

商品描述

The text is divided into three main parts: unconstrained optimization, constrained optimization, and linear programming. The first part addresses unconstrained optimization in single-variable and multivariable functions, introducing key algorithms such as steepest descent, Newton, and quasi-Newton methods.The second part focuses on constrained optimization, starting with linear equality constraints and extending to more general cases, including inequality constraints. It details optimality conditions, sensitivity analysis, and relevant algorithms for solving these problems.The third part covers linear programming, presenting the formulation of LP problems, the simplex algorithm, and sensitivity analysis. Throughout, the text provides numerous applications to data science, such as linear regression, maximum likelihood estimation, expectation-maximization algorithms, support vector machines, and linear neural networks.

商品描述(中文翻譯)

本書的內容分為三個主要部分：無約束優化、有約束優化和線性規劃。第一部分探討單變數和多變數函數的無約束優化，介紹了關鍵算法，如最速下降法（steepest descent）、牛頓法（Newton）和準牛頓法（quasi-Newton methods）。第二部分專注於有約束優化，從線性等式約束開始，擴展到更一般的情況，包括不等式約束。它詳細說明了最優條件、靈敏度分析以及解決這些問題的相關算法。第三部分涵蓋線性規劃，介紹了線性規劃問題的公式化、單純形算法（simplex algorithm）和靈敏度分析。在整個過程中，本書提供了許多數據科學的應用，例如線性回歸、最大似然估計（maximum likelihood estimation）、期望最大化算法（expectation-maximization algorithms）、支持向量機（support vector machines）和線性神經網絡（linear neural networks）。

類似商品