Fig. 4. The Application of the Network
When the label has the multi-attribute[4], we can use this network to generate different things. My another paper is about multi-attribute label.
Conclusion
1.The network achieves the intended function through inverse function. The abstract network can classify and regress, the concrete network can generate input from concept or label.
2.We can change the input 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.
References
  1. Ian Goodfellow, Yoshua Bengio, Aaron Courville, “Deep Learning”, MIT Press, 2016,pp 67-69
  2. Andrew L, Maas, Awni Y, Hannum and Andrew Y. Ng, “Rectified Nonlinearities Improve Neural Network Acoustic Models”, ICML, 2013.
  3. Tensorflow Tutorials and Apis,https://tensorflow.google.cn/learn last accessed 2020/3/20
  4. Jinxin Wei, Qunying Ren “Multi-attribute Recognition,the Key to Universal Neural Network”. unpublished