Computer Science Department @ Princeton University
Deep Learning is the fastest growing area of Machine Learning. This core technique has enabled the latest breakthroughs in computer vision, speech recognition, robotics, natural language processing, and artificial intelligence. It applies neural networks with many layers to large datasets in order to teach computers how to solve perceptual problems, such as detecting recognizable concepts in data, translating or understanding natural languages, interpreting information from input data, and more. Learned models have also been applied for synthesis, for example generating abstract art. In 2016 a Google DeepMind team stunned the acoustic research community with the introduction of WaveNet, a deep learning model that can generate raw audio waveforms containing plausible human speech and even music. Following this work, this seminar will explore applications of deep learning to audio synthesis. Students in the seminar will focus on developing practical deep learning algorithms for modeling and synthesizing audio waveforms.
Students: There are no prerequisites for this seminar aside from the foundational courses for CS majors (126, 217, 226). Like all IW seminars, it is targeted at students starting a first semester of independent work. Some students may have some background in machine learning, or deep learning, or signal processing, or audio, or music – and though not required, this is welcome (especially helpful for getting the rest of us up to speed quickly).
Teams: Students are welcome to work in pairs in the spirit of the whole being greater than the sum of its parts.
Infrastructure: For your projects you can use any software and any dataset you choose. However, we will provide a standard default software platform to ensure that eveyone has a working framework from the beginning. Moreover we will offer several interesting datasets that students may choose to work on.
Milestones: The milestones for this seminar track with the overall steps and deadlines for all independent work projects during the semester. We will also set more fine-grained goals within our seminar.
Resources: Here is a list of resources that may help students get up to speed on deep learning.