Framework for the Interactive Learning of Artificial Neural Networks
We propose framework for interactive learning of artificial neural networks. In this paper we study interaction during training of visualizable supervised tasks. If activity of hidden node in network is visualized similar way as are network outputs, human observer might deduce the effect of this particular node on the resulting output. We allow human to interfere with the learning process of network, thus he or she can improve the learning performance by incorporating his or her lifelong experience. This interaction is similar to the process of teaching children, where teacher observes their responses to questions and guides the process of learning. Several methods of interaction with neural network training are described and demonstrated in the paper.
KeywordsNeural Network Hide Layer Human Observer Multilayer Perceptron Individual Neuron
Unable to display preview. Download preview PDF.
- 5.Edlund, M., Caudel, T.: Realtime visualization of the learning processes in the lapart neural architecture as it controls a simulated autonomous vehicle. Proceedings of the International Joint Conference on Neural Networks 3, 41 (2000)Google Scholar
- 6.Streeter, M., Ward, M., Alvarez, S.A.: Nvis: An interactive visualization tool for neural networks. Proceedings of SPIE Symposium on Visual Data Exploration and Analysis 4302(8), 234–241 (2001)Google Scholar
- 9.Tzeng, F.Y., Ma, K.L.: Opening the black box - data driven visualization of neural networks. In: Proceedings of IEEE Visualization 2005 Conference, pp. 383–390 (2005)Google Scholar
- 11.Užák, M.: Visualization and interaction in the process of neural network learning. Master’s thesis, Technical university of Košice (2005) (in Slovak)Google Scholar