Аннотация
We show how to use "complementary priors" to eliminate the explaining-away effects that make inference difficult in densely connected belief nets that have many hidden layers. Using complementary priors, we derive a fast, greedy algorithm that can learn deep, directed belief networks one layer at a time, provided the top two layers form an undirected associative memory. The fast, greedy algorithm is used to initialize a slower learning procedure that fine-tunes the weights using a contrastive version of the wake-sleep algorithm. After fine-tuning, a network with three hidden layers forms a very good generative model of the joint distribution of handwritten digit images and their labels. This generative model gives better digit classification than the best discriminative learning algorithms. The low-dimensional manifolds on which the digits lie are modeled by long ravines in the free-energy landscape of the top-level associative memory, and it is easy to explore these ravines by using the directed connections to display what the associative memory has in mind.
![Expert Twisted [Event-Driven and Asynchronous Programming with Python]](https://www.rulit.me/data/programs/images/expert-twisted-event-driven-and-asynchronous-programming-wit_606923.jpg "Twisted is a powerful, well-tested, and mature concurrent networking library and framework. As we’ll see in this book, many projects and individuals have used it to great effect for more than a decade.
At the same time, Twisted is large, complicated, and old. Its lexicon teems with strange names,...")
![The Art of Computer Programming: Volume 3: Sorting and Searching [2nd Edition]](https://www.rulit.me/data/programs/images/the-art-of-computer-programming-volume-3-sorting-and-searchi_568628.jpg "The bible of all fundamental algorithms and the work that taught many of today")
Комментарии к книге "A Fast Learning Algorithm for Deep Belief Nets"