An auto-encoder which can be split into two parts is designed. The two parts can work well separately. The top half is an abstract network which is trained by supervised learning and can be used to classify and regress. The bottom half is a concrete network which is accomplished by inverse function and trained by self-supervised learning. It can generate the input of abstract network from concept or label. It is tested by tensorflow and mnist dataset. The abstract network is like LeNet-5. The concrete network is the inverse of the abstract network. A picture can change to label that is compression, then change it back from label that is decompression. So compression and decompression can be realized by the autoencoder. Through test, the absolute function can do generation task well, while the leaky relu function can do classification task well. Lossy compression can be achieved by abstract network and concrete network with absolute function. With one-hot encoding, the compression ratio is 5.1% when decompression quality is good. With binary encoding, the compression ratio is 2% when decompression quality is ok. The mean PNSR of five pictures is 19.48dB. When jump connection and negative feedback are used, the decompression performance is good. The mean PNSR of five pictures is 29.86dB. The compression ratio is 56.1%.

An auto-encoder which can be split into two parts is designed. The two parts can work well separately. The top half is an abstract network which is trained by supervised learning and can be used to classify and regress. The bottom half is a concrete network which is accomplished by inverse function and trained by self-supervised learning. It can generate the input of abstract network from concept or label. It is tested by tensorflow and mnist dataset. The abstract network is like LeNet-5. The concrete network is the inverse of the abstract network. Through test, encryption and decryption can be achieved by abstract network and concrete network add jump connect and negative feedback with absolute function. The parameter of DNN is secondary key, the architecture of DNN is primary key. Secondary key can be shared by all the people, primary key can be shared by sender and receiver. The key can be generated by training the DNN. Through analysis, it is safe for most situations.

In order to solve the large classes problem, binary encoding for label is proposed. Through test on mnist dataset for classification task, The classification accuracy with binary encoding is almost equal to one-hot encoding. Large classes can be solved by binary encoding. Lossy compression can be achieved by abstract network and concrete network with absolute function. With one-hot encoding, the compression ratio is 5.1% when decompression quality is good. With binary encoding, the compression ratio is 2% when decompression quality is ok. When jump connection and negative feedback are used, the decompression performance is good.

Inspired by my another paper[3], neural network can detect location and size,so it can do object detection task. I redesign the mnist dataset, to produce different location and different size. Through test on new mnist dataset, the rectangle of number is all right both on absolute function and relu function. The classification accuracy is 97%-98%, the accuracy of location of rectangle is 99%-100%. It is easy for location task than for classification task. It is easy to distinguish black and white for one object, but RGB picture is harder to distinguish. Because the role of neural network is like a function. The location task can achieved by neural network as long as the problem can be expressed by a function. The task can be done as long as there is enough data and enough network complexity. The whole task is suit to study object detection.When we use lidar, the picture we got is like the gray picture listed before, we just to classify and locate the way we can drive through, so one object should be classified and located, the accuracy is high.

Inspire by nature world mode, a activation function is proposed. It is absolute function.Through test on mnist dataset and fully-connected neural network and convolutional neural network, some conclusions are put forward. The line of accuracy of absolute function is a little shaken that is different from the line of accuracy of relu and leaky relu. The absolute function can keep the negative parts as equal as the positive parts, so the individualization is more active than relu and leaky relu function. In order to generalization, the individualization is the reason of shake, the accuracy may be good in some set and may be worse in some set. The absolute function is less likely to be over-fitting. The batch size is small, the individualization is clear, vice versa. Through test on mnist and autoencoder, It is that the leaky relu function can do classification task well, while the absolute function can do generation task well. Because the classification task need more universality and generation task need more individualization. The pleasure irritation and painful irritation is not only the magnitude differences, but also the sign differences, so the negative part should keep as a part. Stimulation which happens frequently is low value, it is showed around zero in figure 1 . Stimulation which happens accidentally is high value, it is showed far away from zero in figure 1. So the high value is the big stimulation, which is individualization.

To achieve the recognition of multi-attribute of object, I redesign the mnist dataset, change the color, size, location of the number. Meanwhile, I change the label accordingly. The deep neural network I use is the most common convolution neural network. Through test, we can conclude that we can use one neural network to recognize multi-attribute so long as the attribute difference of objects can be represented by functions. The Concrete network (generation network) can generate the output which the input rarely contained from the attributes the network learned. Its generalization ability is good because the network is a continuous function. Through one more test, we can conclude that one neural network can do image recognition, speech recognition, nature language processing and other things so long as the output node and the input node and more parameters add into the network. The network is universal so long as the network can process different inputs. By proof, fully connected network can do what convolution neural network and recurrent neural network do, so fully connected network is the universal network. The phenomenon of synesthesia is the result of multi-input and multi-output. Connection in mind can realize through the universal network and sending the output into input. Connection in mind is the key of creativity, synesthesia is the assistant.

According to kids’ learning process, an auto-encoder which can be split into two parts is designed. The two parts can work well separately. The top half is an abstract network which is trained by supervised learning and can be used to classify and regress. The bottom half is a concrete network which is accomplished by inverse function and trained by self-supervised learning. It can generate the input of abstract network from concept or label. The network can achieve its intended functionality through testing by mnist dataset and convolution neural network. Round function is added between the abstract network and concrete network in order to get the representative generation of class. The generation ability can be increased by adding jump connection and negative feedback. At last, the characteristics of the network is discussed. The input can be changed to any form by encoder and then change it back by decoder through inverse function. The concrete network can be seen as the memory stored by the parameters. Lethe is that when new knowledge input, the training process makes the parameters change. At last, the application of the network is discussed. The network can be used for logic generation through deep reinforcement learning. The network can also be used for language translation, zip and unzip, encryption and decryption, compile and decompile, modulation and demodulation.