Inductive Biases in Machine Learning for Robotics and Control
Lutter, Michael
- 出版商: Springer
- 出版日期: 2024-08-02
- 售價: $4,640
- 貴賓價: 9.5 折 $4,408
- 語言: 英文
- 頁數: 119
- 裝訂: Quality Paper - also called trade paper
- ISBN: 3031378342
- ISBN-13: 9783031378348
-
相關分類:
機器人製作 Robots、Machine Learning
海外代購書籍(需單獨結帳)
相關主題
商品描述
One important robotics problem is "How can one program a robot to perform a task"? Classical robotics solves this problem by manually engineering modules for state estimation, planning, and control. In contrast, robot learning solely relies on black-box models and data. This book shows that these two approaches of classical engineering and black-box machine learning are not mutually exclusive. To solve tasks with robots, one can transfer insights from classical robotics to deep networks and obtain better learning algorithms for robotics and control. To highlight that incorporating existing knowledge as inductive biases in machine learning algorithms improves performance, this book covers different approaches for learning dynamics models and learning robust control policies. The presented algorithms leverage the knowledge of Newtonian Mechanics, Lagrangian Mechanics as well as the Hamilton-Jacobi-Isaacs differential equation as inductive bias and are evaluated on physical robots.
商品描述(中文翻譯)
一個重要的機器人學問題是「如何編程讓機器人執行任務?」傳統的機器人學通過手動設計狀態估計、規劃和控制模組來解決這個問題。相對而言,機器人學習則完全依賴黑箱模型和數據。本書顯示,傳統工程和黑箱機器學習這兩種方法並不是互相排斥的。為了解決機器人的任務,可以將傳統機器人學的見解轉移到深度網絡上,從而獲得更好的機器人學習算法和控制方法。為了強調將現有知識作為歸納偏見納入機器學習算法能提高性能,本書涵蓋了學習動態模型和學習穩健控制策略的不同方法。所提出的算法利用牛頓力學、拉格朗日力學以及哈密頓-雅可比-伊薩克斯微分方程作為歸納偏見,並在物理機器人上進行評估。