Record why RNN can record previous historical information, how to reflect it from the formula?
Then first introduce why the ordinary neural network can not remember the previous historical information, and then lead to the corresponding ability of the RNN, because if the ordinary neural network can record the previous historical information, then there will be no RNN thought.
1 common neural network (MLP)First of all, we have a task, which is to perform part-of-speech tagging. There are two training data below.
He confessed to me that I think his confession is not sincere enough.
The correct part of speech is:
Then the training data is sent to the neural network for training, such as the first data "he/r", the neural network learning a mapping of "he->r", the second data "to /p", the neural network learning one The mapping to "--p", so that the training data has been learned, updated to the final parameters, and thus learn the model, but the problem is coming.
The example of the study is as follows:
In the above training data, the part of speech is not unique. For example, the word "confession" is used as the verb v in the phrase "he confesses to me". In the sentence "I think his confession is not sincere enough" As a noun n in words, it is a mess for the neural network.
All of a sudden neural network should learn that "confession" is a verb. At once, it is necessary to learn that "confession" is a noun. The neural network is also very innocent. It has no ability to deal with what should be called "confession" as a noun and what will fall. "Expression" is judged as a verb because the neural network does not learn the surrounding context. The data fed to the neural network is not linked to the previous data.
So we need a network that can remember the historical information of the past. For example, in the first sentence, when I encountered the word expression, I know that the word in front of him is "I" / pronoun, then the confession behind the pronoun The probability of being a verb is much greater than the confession of a noun. Of course, RNN can also see several words in front of him. In theory, rnn can memorize any word in front of the current word.
In the second sentence, when the word "confession" is encountered, our network can know that the word in front of him is ""/auxiliary, then the probability of "confession" behind the auxiliary word as a noun is much greater than The confession of the verb.
So we hope that we can have a network to predict the current task, to remember the previous knowledge to help the current task to complete, so that RNN will debut, and some small partners may say that it has many problems, such as not being able to remember for a long time. But this article does not introduce, but in any case, RNN provides the possibility to solve this problem.
2 cyclic neural network record history information RNNLet me introduce RNN first.
First look at a simple cyclic neural network, which consists of an input layer, a hidden layer, and an output layer:
I don't know if the beginners can understand this picture. Anyway, when I first started learning, it is very aggressive. Each node represents a value input, or a layer of vector node collection. How to hide the layer? Can connect to yourself, and so on. These figures are a more abstract picture.
We now understand that if the circle with the arrow above W is removed, it becomes the most common fully connected neural network. x is a vector that represents the value of the input layer (the circle that is not drawn to represent the neuron node); s is a vector that represents the value of the hidden layer (here the hidden layer draws a node, you can also imagine this A layer is actually a plurality of nodes, and the number of nodes is the same as the dimension of the vector s);
The game keyboard requires a strong sense of key paragraphs,so as to produce a special feel suitable for game entertainment,and achieve a good experience for players in the game. The keyboard is one of the indispensable computer peripherals in our life. The previous keyboard has been used as an input device for office typing.With the emergence of computer games, the keyboard has gradually become a peripherals for games. Due to the relatively low performance of early computers,the way of computer games in this period is also relatively simple,ordinary office and home keyboard can completely meet the needs of players.However, with the continuous upgrading of hardware performance,the fun and complexity of games are also increasing, and the requirements for keyboard performance are also getting higher and higher,so the professional Gaming Keyboard was born.
As one of the most important peripherals for playing games,the tactile feel of the gaming keyboard is the most important reference.Tactile sensation is mainly determined by the strength of the key resistance degree.To judge the feel of a keyboard,it will be tested from whether the key elastic is moderate,whether the key force is uniform,whether the key cap is loose or shaken,and whether the key process is appropriate. Although different players have different requirements for the stretch and range of the keys,the most commonly accepted peripherals for playing games are the Black Axis Gaming Keyboard Mechanical.
The appearance includes the color and shape of the keyboard.A nice and stylish gaming keyboard will add a lot of color to your desktop,while a stuffy Wireless Gaming Keyboard will make you feel dull when playing games. Therefore,for the keyboard,as long as the player feels beautiful like, practical can be.
Gaming Keyboard,Led Gaming Keyboard,Keyboard Gaming Mobile,Optical Mechanical Keyboard
Henan Yijiao Trading Co., Ltd , https://www.yjusbhubs.com