Deep Learning

Harvard Extension School

CSCI E-89

Section 1

CRN 16392

View Course Details
In this course, students master the most important neural network and deep learning concepts and techniques and start applying them productively in modern artificial intelligence (AI) workplace. Deep learning is the primary technique for data analysis and the solution for many complex problems in natural sciences, linguistics, and engineering. We use deep learning for image classification, manipulation, and generation; speech recognition and synthesis; natural language translation; sound and music manipulation and generation; and navigation of self-driving cars. Students master several key deep learning architectures, such as convolutional neural networks (CNNs), long short-term memory networks (LSTMs), autoencoders (AEs), variational autoencoders (VAEs), generative adversarial networks (GANs), transformers with attention, and graph neural networks (GNNs). Students master the most essential skills for the efficient use of large language model (LLM)-based applications such as ChatGPT and DALL*E. The course starts with a review of the theoretical foundations of neural networks approach to machine learning including auto-differentiation and back-propagation. The emphasis is on practical applications of deep learning APIs Keras (a package within TensorFlow 2.x framework) and PyTorch.

Instructor Info

Zoran B. Djordjević, PhD

Senior Enterprise Architect, Nishava, Inc.


Rahul B. Joglekar, BSc

Enterprise Technical Architect, Point32Health


Meeting Info

F 6:30pm - 8:30pm (9/2 - 12/20)

Participation Option: Online Asynchronous or Online Synchronous

In online asynchronous courses, you are not required to attend class at a particular time. Instead you can complete the course work on your own schedule each week.

Deadlines

Last day to register: August 28, 2025

Additional Time Commitments

Optional sections Saturdays, 1-2 pm.

Prerequisites

Proficiency with Python. We assume no familiarity with Linux and introduce all essential Linux features and commands. Students need access to a computer with a 64-bit operating system and at least 8 GB of RAM. Having a machine with NVIDIA card is a plus but not required. All complex examples given as assignments could be run on Google Collaboratory.

Notes

This course meets via web conference. Students may attend at the scheduled meeting time or watch recorded sessions asynchronously. Recorded sessions are typically available within a few hours of the end of class and no later than the following business day. See minimum technology requirements.

All Sections of this Course

CRN Section # Participation Option(s) Instructor Section Status Meets Term Dates
34723 1 Online Asynchronous, Online Synchronous Dmitry Kurochkin Open MW 6:30pm - 9:30pm
Jun 23 to Aug 8
16392 1 Online Asynchronous, Online Synchronous Team Taught Open F 6:30pm - 8:30pm
Sep 2 to Dec 20