A survey class of neural network implementation and applications. Topics include optimization - stochastic gradient descent, adaptive and 2nd order methods, normalization; convolutional neural networks - image processing, classification, detection, segmentation; recurrent neural networks - semantic understanding, translation, question-answering; cross-domain applications - image captioning, vision, and language.
There will be a final project worth 20% of your final grade. The project can be done individually or in teams.
For your final project you should explore any topic you are interested in related to deep learning. This could involve training a model for a new task, building a new dataset, improving deep models in some way and testing on standard benchmarks, etc. You project should probably involve some implementation, some data, and some training. The amount of effort and time should be approximately 2 homework assignments.
Your final project presentation will be a website describing your project, and a 2-3 minute video. This summary should mention the problem setup, data used, techniques, etc. It should include a description of which components were from preexisting work (i.e. code from github) and which components were implemented for the project (i.e. new code, gathered dataset, etc).