Deep Learning Techniques

Deep Learning Techniques (MDS-301) is a core course for M.Sc. Data Science – III Semester that provides a comprehensive foundation in artificial neural networks and modern deep learning methods. The course covers neural network models, activation functions, ANN architectures, supervised and unsupervised learning, perceptron, gradient descent, backpropagation, Hebbian and competitive learning, Self-Organizing Maps, Radial Basis Function Networks, and reinforcement learning. It further introduces advanced architectures including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), LSTMs, Generative Adversarial Networks (GANs), Deep Belief Networks, Markov Decision Processes, and Hidden Markov Models. The course emphasizes both theoretical foundations and practical applications, enabling students to understand, develop, and apply deep learning models to real-world data science problems.

REFERENCES

1. Haykin, S. (1994). Neural Networks: A Comprehensive Foundation. New York: Macmillan Publishing. A comprehensive book and contains a great deal of background theory.
2. Yagnanarayana, B. (1999): Artificial Neural Networks, PHI.
3. Bart Kosko (1997): Neural Networks and Fuzzy Systems, PHI.
4. Jacek M. Zurada (1992): Artificial Neural Systems, West Publishing Company.
5. Carling, A. (1992). Introducing Neural Networks. Wilmslow, UK: Sigma Press.
6. Fausett, L. (1994). Fundamentals of Neural Networks. New York: Prentice Hall.
7. Box and Jenkins: Time Series Analysis, Springer.
8. Brockwell, P.J., and Davis, R.A.: Time Series: Theory and Methods (Second Edition). Springer-Verlag.