A project-based introduction to deep learning architectures, algorithms, and applications. Includes multilayer perceptrons, convolutional neural networks, recurrent neural networks, and transformers; theory and implementation of learning algorithms, training and tuning procedures; applications to computer vision and natural language processing.
We will focus on classification, regression, and unsupervised learning problems with artificial neural networks (ANNs). ANNs are algorithms we train to map input datapoints (could be text, images, audio/video files, database entries) to desired targets (numbers, categories, patterns, etc.). Through this training, the computer learns to classify (or estimate) targets for unknown data points.
Can a computer classify a picture of an animal as a cat? Can it differentiate between hip-hop and jazz music? Can it interpret the sentiment behind text? Can it generate synthetic data?