Students will build deep learning prediction models of different complexities, from simple linear logistic regression to major categories of neural networks including convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory (LSTMs), and gated recurrent units (GRUs).

By the end of the course, students will be proficient in best practices of using standard machine learning frameworks such as Pytorch, TensorFlow and Keras, and using datasets for solving common machine learning problems.

The class prepares students to pursue a career in data sciences and AI model development.

Learning Outcomes
At the conclusion of the course, you should be able to

• Use common deep learning architectures such as CNN and RNN
• Discuss the significance of hyperparameters in the architectures
• Prepare data for deep learning using Pandas and NumPy, the de facto standard for data prep in Python
• Write scalable code and develop machine learning models that can be used to train deep learning architectures on real-world business problems
• Debug and understand the inner working of deep learning architectures

Topics Include

• Deep learning with standard machine learning frameworks including TensorFlow, Keras and Pytorch
• Multilayer perceptrons
• Advanced multilayer perceptrons
• Convolutional neural networks
• Image processing CNN architectures
• Recurrent neural networks
• RNN - prediction with multilayer perceptron
• RNN - prediction with long short term memory networks

Note(s): Students are required to bring laptops for the classroom and work with Python3/ Jupyter Notebook environment.

Skills Needed: Moderate level of computer programming ability in Python, comfortable with an editor, familiarity with command-line operations on a laptop, and a basic understanding of Machine Learning models.