Artificial Intelligence Applications in Engineering (MUH-920), week 10 of 14: interactive lab
Prof. Dr. Utku Kose, Süleyman Demirel University
Part A shows a single neuron with adjustable weights, bias and activation over a two-dimensional input space. Part B trains a network with one or two hidden layers on circles, XOR, two moons or a spiral, with controls for width, depth, activation and learning rate [1]. Part C compares gradient descent, momentum and Adam on a narrow valley of a loss surface [2].
Six questions with instant feedback. Rate your confidence before checking each answer.
Answers are saved in this browser only. The export creates a Markdown learning log for your portfolio.
After the export, continue with the discipline challenge and the weekly task in the week overview.
[1] Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323(6088), 533-536. https://doi.org/10.1038/323533a0
[2] Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations (ICLR 2015). https://arxiv.org/abs/1412.6980
[3] Rosenblatt, F. (1958). The perceptron: A probabilistic model for information storage and organization in the brain. Psychological Review, 65(6), 386-408. https://doi.org/10.1037/h0042519
[4] Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals and Systems, 2(4), 303-314. https://doi.org/10.1007/BF02551274
[5] Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359-366. https://doi.org/10.1016/0893-6080(89)90020-8
[6] Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(56), 1929-1958.