Opening book details…
Can I read Differentiation and Adjoints on EtoBox?
Differentiation and Adjoints is a book available to read on EtoBox.
What is Differentiation and Adjoints about?
## Differentiation and Adjoints In this chapter, we describe efficient calculation of gradients for certain structured functions, particularly those that arise in the training of deep neural networks (DNNs). Such functions share some features with objective functions that arise in such applications as data assimilation and control, in both of which the optimization problem is integrated with a model of a dynamic process, one that evolves in time or proceeds by stages. In deep learning, the progressive transformation of each item of data as it moves through the layers of the network is akin to a dynamic process.
- Publisher
- Cambridge University Press
- Published
- 2022
- Language
- EN
Similar books
- Automatic Differentiation in Deep Learning — Nikhil Ketkar; Jojo Moolayil (2021)
- Automatic Differentiation of Algorithms : from Simulation to Optimization ; [selected Papers from the Third International Conference on Automatic Differentiation] ; Includes CD-ROM — George Corliss; Christele Faure; Andreas Griewank; Laurent Hascoet; Uwe Naumann (2013)
- Transdifferentiation : Flexibility in Cell Differentiation — Tokinde S Okada (1991)
- Tangent Linear Modeling and Adjoints — Steven J. Fletcher (2023)
- Sex Differentiation and Development — C. R. AUSTIN (1960)
- Epigenetics, Stem Cells, Cellular Differentiation, and Associated Neurological Disorders and Brain Cancer — Bhairavi Srinageshwar; Gary L. Dunbar; Julien Rossignol (2023)