Deep Belief Nets in C++ and CUDA C: Volume III: Convolutional Nets (Volume 3) (Paperback)

Deep belief nets are one of the most exciting recent developments in artificial intelligence. The structure of these elegant models is much closer to that of human brains than traditional neural networks; they have a ‘thought process’ that is capable of learning abstract concepts built from simpler primitives. A typical deep belief net can learn to recognize complex patterns by optimizing millions of parameters, yet this model can still be resistant to overfitting. This book presents the essential building blocks of a common and powerful form of deep belief net: convolutional nets. These models are especially useful for image processing applications. At each step the text provides intuitive motivation, a summary of the most important equations relevant to the topic, and concludes with highly commented code for threaded computation on modern CPUs as well as massive parallel processing on computers with CUDA-capable video display cards. Source code for all routines presented in the book, and the executable CONVNET program which implements these algorithms, are available for free download from the author’s website. Source code for the complete CONVNET program is not available, as much of it is highly specialized Windows interface code. Readers are responsible for writing their own main program, with all interface routines. You may freely use all of the core convolutional net routines in this book, as long as you remember that it is experimental code that comes with absolutely no guaranty of correct operation.

深度信念網絡是人工智慧領域中最令人興奮的最新發展之一。這些優雅模型的結構比傳統神經網絡更接近人類大腦，它們具有一種能夠從更簡單的基本元素中學習抽象概念的“思考過程”。一個典型的深度信念網絡可以通過億萬參數的優化來學習識別複雜模式，但這種模型仍然可以抵抗過度擬合。本書介紹了一種常見且強大的深度信念網絡結構的基本構建塊：卷積網絡。這些模型對於圖像處理應用特別有用。在每一步中，本書提供直觀的動機，總結了與該主題相關的最重要的方程式，並以高度註釋的代碼結束，該代碼可在現代CPU上進行線程計算，也可在具有CUDA兼容顯示卡的計算機上進行大規模並行處理。本書中介紹的所有例程的源代碼以及實現這些算法的可執行CONVNET程序都可以從作者的網站免費下載。完整的CONVNET程序的源代碼不可用，因為其中很多是高度專門化的Windows界面代碼。讀者需要自己編寫自己的主程序，包括所有界面例程。您可以自由使用本書中的所有核心卷積網絡例程，但請記住，這是實驗性代碼，並不保證正確運行。