Classification Networks Mạng lưới thần kinh có thể được huấn luyện để bản đồ một không gian đầu vào bất kỳ loại không gian đầu ra. Ví dụ, trong chương trước, chúng tôi khám phá một bản đồ homomorphic, trong đó các đầu vào và đầu ra không gian là như nhau, và các mạng lưới đã được dạy để đưa ra dự đoán hoặc interpolations trong không gian đó. Một loại hữu ích của bản đồ phân loại, trong đó vector đầu vào được ánh xạ vào một trong các lớp học N. Một mạng lưới thần kinh có. | 7. Classification Networks Neural networks can be taught to map an input space to any kind of output space. For example in the previous chapter we explored a homomorphic mapping in which the input and output space were the same and the networks were taught to make predictions or interpolations in that space. Another useful type of mapping is classification in which input vectors are mapped into one of N classes. A neural network can represent these classes by N output units of which the one corresponding to the input vector s class has a 1 activation while all other outputs have a 0 activation. A typical use of this in speech recognition is mapping speech frames to phoneme classes. Classification networks are attractive for several reasons They are simple and intuitive hence they are commonly used. They are naturally discriminative. They are modular in design so they can be easily combined into larger systems. They are mathematically well-understood. They have a probabilistic interpretation so they can be easily integrated with statistical techniques like HMMs. In this chapter we will give an overview of classification networks present some theory about such networks and then describe an extensive set of experiments in which we optimized our classification networks for speech recognition. . Overview There are many ways to design a classification network for speech recognition. Designs vary along five primary dimensions network architecture input representation speech models training procedure and testing procedure. In each of these dimensions there are many issues to consider. For instance Network architecture see Figure . How many layers should the network have and how many units should be in each layer How many time delays should the network have and how should they be arranged What kind of transfer function should be used in each layer To what extent should weights be shared Should some of the weights be held to fixed values Should output units be .
