Deep Neural Networks
Course Description
Deep learning is the engine behind many of today’s data-driven products and services – ranging from financial services to health and from self-driving cars to security. This introductory course will help you understand how and why deep learning has proven to be so effective, and how its usage proliferated over the past several years. Course content will provide you with the foundational knowledge necessary to understand the methods used in deep learning, and to be able to start to use it within your academic studies or professional life.
This course provides an introduction to Deep Learning. It introduces core courses of the domain in an easy to understand and accessible manner that does not presume any previous knowledge of or experience with programming, statistics, or deep learning itself. Students will be led to develop fundamental knowledge and skills necessary for further study of deep learning.
The teaching takes place online in small and interactive groups. The Friday lectures introduce core concepts and offer the students the opportunity to ask questions and interact with the course lead. The Saturday TA seminars recapitulate the core lessons of the week and give the students the opportunity to re-visit the material they found particularly challenging. Students will be supported throughout the duration of the course, both in lectures and in TA seminars, by a Chinese speaking Teaching assistant. Upon completion, students will have the opportunity to discuss their performance in a one-on-one session with the course lead. All lectures and TA seminars will be recorded and made availible to registered students during their module’s duration.
All students who complete the course will receive a course completion certificate issued by the Cambridge AI Academy, and signed by Nicholas Lane. Furthermore, students will be eligible to receive a personalized letter of performance written by Dr Nicholas Lane, suitable for use as support for applications to further study. This letter will comment on the students demonstrated abilities, skills, and progress made during the course.
Course lead Nicholas D. Lane
Professor at the Department of Computer Science and Technology at the University of Cambridge where he leads the Machine Learning Systems Lab. Prior to joining Cambridge, Dr Lane was an Associate Professor at the University of Oxford (2017 to 2020) and Senior Lecturer at the University of London (2016 to 2017). Nicholas also has more than 10 years of experience in industrial research. Alongside his academic position, he is currently a Director at the Samsung AI Centre in Cambridge. Previously, he has been a Principal Scientist at Nokia Bell Labs and a Lead Researcher at Microsoft Research in Beijing.
Course syllabus
Lecture 1 (free, week 0) will introduce background material that explains what made Deep Learning the juggernaut it is today and the forces that are likely to shape its evolution in the future. It is free to attend and held two weeks before the primary course material begins.
Lecture 2 (week 1) will introduce the core building blocksthat power deep models. Traditional architectures, convolutional and recurrent networks, will be covered. Applications in vision and natural language processing will be discussed amd code examples will be presented.
Lecture 3 (week 2) will introduce the advanced building blocks that power modern deep models. Residual connections, depth-wise separability, channel shuffle, and attention will be covered. Applications in vision and natural language processing will be discussed amd code examples will be presented.
Lecture 4 (week 3) will examine the art of training deep neural networks. First, a solid understanding of deep learning parameter learning will be derived from core statistical concepts. Then, a step-by-step guide to successfully training a deep model will be built on top of this foundation.
Lecture 5 (week 4) will explore the fundamentals of deep learning computational resource management. The chief impediment to further deep learning adoption is its dependence on expensive large-scale compute. This lecture will tackle this dependence both from the vantage point of computer systems as well as the algorithms themselves.
Lecture 6 (week 5) will take a deep dive into the hottest topics in the bleeding-edge deep learning literature. The most exciting and influential deep learning papers will be presented, and their methods will be analysed in detail. The students will learn how to critically assess and learn from the best of the best papers. in Deep Learning.
Lecture 7 (week 6) will be structured as a student-led reading group. The students will be given the opportunity to put the knowledge they learned over the preceding 5 weeks to test and practice their academic communication, presentation, and research evaluation skills by delivering concise summaries of the most important papers in deep learning.
The course will conclude with a multiple-choice exam conducted in real-time online and a 10-minute personalized one-to-one feedback session. Students will be provided with a certificate of accomplishment and any number of letters of recommendation based on their in-course performance for up to 24 months after finishing the course.