LibGuides: CSC453: Module 3 Resources

Module 3 Required Resources

Neural Network Design
Hagan, M. T., Demuth, H. B., Beale, M. H., & De Jesús, O. (2014). Neural network design (2nd ed.). Martin Hagan.
These chapters build core vocabulary on generalization and dynamic learning. The readings introduce estimating generalization error, early stopping, regularization, Bayesian regularization, and the connection between early stopping and regularization, as well as layered digital dynamic networks, BPTT versus RTRL, and stability notes. They support MLO 1 to MLO 4 and prepare you for the Module 3 quiz and assignment.
Chapter 13 – Generalization: estimating generalization error, early stopping, regularization, Bayesian regularization, and the early stopping to regularization connection.
Chapter 14 – Dynamic Networks: layered digital dynamic networks, principles of dynamic learning, BPTT versus RTRL, and stability notes.

scikit-learn Model Selection Guide
scikit-learn. (n.d.). Model selection: Choosing estimators and their parameters. https://scikit-learn.org/stable/modules/cross_validation.html
Official scikit-learn documentation covering model validation and cross-validation techniques.
Understanding LSTM Networks
Olah, C. (2015, August). Understanding LSTM networks. https://colah.github.io/posts/2015-08-Understanding-LSTMs/
Christopher Olah’s intuitive visual explanation of how Long Short-Term Memory (LSTM) networks work.
Statsmodels Time Series Quickstart
Seabold, S., & Perktold, J. (n.d.). Time series analysis — statsmodels documentation. https://www.statsmodels.org/stable/tsa.html
Statsmodels quickstart guide to time series analysis, useful for building baseline models.