AT500 – Applied Artificial Intelligence & Deep Learning

The objective of this course is to introduce students to artificial intelligence with intelligent agents, knowledge representation, reasoning, machine learning, and deep learning. It will then present methods for language analysis and computer vision.

The first objective will be to learn the basic ideas in Artificial Intelligence behind the design of intelligent agents and knowledge representation with logical/probabilistic reasoning for real-world data/problems. The latter include search, games, machine learning, and constraint satisfaction.

The next objective will be to review basic ideas of design and analysis of efficient algorithms, including searching, graph algorithms, and dynamic programming. It will then emphasize on supervised machine learning techniques for regression and classification. The latter using the scikit-learn library.

Then, it is to learn about Deep Learning (DL) for neural networks. In particular, cover datasets, architecture, optimization, training, and performance validation. It will use their implementation with the TensorFlow and Keras deep learning libraries.

Eventually, the course will introduce systems to analyze, understand, and produce language and sequences. In particular, the use of machine and deep learning methods for language modeling. It will examine applications such as information extraction, machine translation, and text generation.

The last topic will be on aspects of computer vision. It will introduce ideas on image processing, filtering, segmentation, object recognition, and detection with deep neural networks.