Introduction to Deep Learning
computer science
about this course
This beginner-level deep learning course is
designed for 3rd- and 4th-year students and can also serve as a fresh
introduction to modern machine learning. It provides a thorough
foundation in core concepts by beginning with classical methods such as
regression and classification, and then building naturally towards shallow
and deep neural networks.
The course begins with regression through a simple curve-fitting
example, introducing key ideas such as error functions, regularization,
and the bias–variance trade-off. Regression and classification are then
presented as single-layer networks, highlighting linear models,
discriminant functions, and generative vs. discriminative classifiers.
These methods motivate the shift from fixed basis functions to learning
representations directly from data, naturally leading to neural networks.
Students are then introduced to deep neural networks, with a focus on
gradient descent, backpropagation, and regularization. The course
concludes with an introduction to Convolutional Neural Networks (CNNs)
and their practical applications
in these programmes
- computer science
elective · 4-year major, 4-year double major, 3-year major, 3-year double major, 4-year minor, 3-year minor
- data science
required · 4-year major, 4-year double major, 3-year major, 3-year double major, 4-year minor, 3-year minor
planning & teaching
before this
No prerequisites.
recently taught by
- Ravi Prakash IyerVisiting Professor, Computer Science
Recorded in 2025 T3, 2024 T1. Instructors may change.