Author: Zen and the Art of Computer Programming
1 Introduction
In recent years, the field of deep learning has become one of the indispensable research hotspots in all walks of life. With the rapid development of deep learning, more and more people began to study, apply and explore it. But to really master and understand deep learning, you need to understand some important principles and related algorithms.
"Deep Learning" (deeplearning.ai), written by Professor LeCun, the originator of the field of deep learning, was written by Wu Enda and others in 2015.
The basic principles and key algorithms of deep learning are described in detail in the book "Popular Neural Networks", and there are corresponding code implementations. By reading this book, you can not only understand the basic knowledge and practical methods of deep learning, but also have a deeper understanding of the working mechanism of deep learning and apply it to actual projects.
2. Explanation of basic concepts and terms
(1) Neuron model
Perceptron model
Perceptron is a linear classifier for binary classification. The input feature vector x is multiplied by a weight vector w to obtain a predicted value o. If o is greater than a certain threshold, it outputs +1, otherwise it outputs -1. The learning strategy of the perceptron is to continuously adjust the weight vector w during the training process, so that the result of the input feature vector x is as close as possible to the expected output y.
Hidden Markov Model (HMM)
Hidden Markov Model (HMM) is used for labeling problems (speech recognition, POS tagging, named entity recognition, machine translation, ima