Deep Learning

Deep Learning

taught by Pragyansmita Nyack


Aim of Course:

In this online course, you will learn about the rapidly evolving field of Deep Learning. The surge in deployed applications based on concepts and methods in this field is an indication of its potential to help fully realize the promise of Artificial Intelligence. At the end of this course you will understand the basic concepts underlying the representations and methods in deep learning and see some applications where deep learning is most effective. You will also gain an appreciation of what kind of problems are most suited for this field and current research trends.

Antiicpated Learning Outcomes:

  • Install and run Tensorflow
  • Build on the available Tensorflow code to run a linear classifier and neural net
  • Explain the characteristics and parameters of a convolutional network
  • Use Tensorflow to configure a convolutional network to recognize handwritten digits
  • Explain the role played by networks based on Long Short Term Memory
  • Explain the considerations involved in implementing both supervised and unsupervised methods in a deep network


This course may be taken individually (one-off) or as part of a certificate program.
Course Program:

WEEK 1: Neural Networks and Optimization

  • Overview
  • Machine Learning Basics
  • Deep Feedforward Networks
  • Basic Optimization Algorithms

WEEK 2: Convolutional Networks and Image Processing

  • Convolutional Networks
  • Practical Methodology
  • Image Applications


WEEK 3: Recurrent Architectures and Language Processing

  • Recurrent Networks
  • Long Short Term Memory
  • Language Applications

WEEK 4: Advanced Topics, Research Trends

  • Autoencoders
  • Representation Learning
  • Deep Generative Models


Homework in this course consists of short answer questions to test concepts, practice with using public-domain tools and some guided exercises that involve freely available data. There is also an end-of-course project.

Deep learning and its place in the learning hierarchy


From Deep Learning by Ian Goodfellow and Yoshua Bengio and Aaron Courville


Deep Learning

Who Should Take This Course:
Data scientists, statisticians, software engineers, technical managers interested in learning about what is deep learning, the state-of-the-art techniques/concepts and the range of applications in which they can be used.
You should have some familiarity with linear algebra, probability and information theory. The course text's chapters 2 and 3 can be read in advance to gain this background. You should also be able to run open-source tools and install them on your personal computer. You should be sufficiently familiar with Python to write programs to read/generate data and process it in appropriate data structures using relevant packages.
Organization of the Course:

This course takes place online at the Institute for 4 weeks. During each course week, you participate at times of your own choosing - there are no set times when you must be online. Course participants will be given access to a private discussion board. In class discussions led by the instructor, you can post questions, seek clarification, and interact with your fellow students and the instructor.

At the beginning of each week, you receive the relevant material, in addition to answers to exercises from the previous session. During the week, you are expected to go over the course materials, work through exercises, and submit answers. Discussion among participants is encouraged. The instructor will provide answers and comments, and at the end of the week, you will receive individual feedback on your homework answers.

Time Requirement:
About 15 hours per week, at times of  your choosing.

Options for Credit and Recognition:
Students come to the Institute for a variety of reasons. As you begin the course, you will be asked to specify your category:
  1. No credit - You may be interested only in learning the material presented, and not be concerned with grades or a record of completion.
  2. Certificate - You may be enrolled in PASS (Programs in Analytics and Statistical Studies) that requires demonstration of proficiency in the subject, in which case your work will be assessed for a grade.
  3. CEUs and/or proof of completion - You may require a "Record of Course Completion," along with professional development credit in the form of Continuing Education Units (CEU's).  For those successfully completing the course,  CEU's and a record of course completion will be issued by The Institute, upon request.
  4. Digital Badge - Courses evaluated by the American Council on Education have a digital badge available for successful completion of the course.  
  5. Other options - Specializations, INFORMS CAP recognition, and academic (college) credit are available for some courses

Specializations are an easy way for you to demonstrate mastery of a specific skill in statistics and analytics. This course is part of the Intelligence and Security Analytics Specialization which teaches statistical and machine learning methods for detecting anomalies, identifying images, and processing data from sensors.

College credit:
Deep Learning has been evaluated by the American Council on Education (ACE) and is recommended for the upper division baccalaureate degree category, 3 semester hours in computer science, machine learning, or artificial intelligence. Note: The decision to accept specific credit recommendations is up to each institution. More info here.
Course Text:

The required text is Deep Learning (MIT Press, 2016) by I. Goodfellow, Y. Bengio and A. Courville.

A freely available html version is available at .

It will be supplemented in the course by other technical papers, also available online.


The assignments will use TensorFlow.
Instructions for installing TensorFlow can be found here:

We recommend installing Python 3 (>= 3.4) and pip3, although
TensorFlow can also be run using Python 2 (>= 2.7) and pip.
It is possible to have both versions of Python on your computer
and switch between them for different purposes.
One advantage of Python 3 is that it uses Unicode natively,
which is often convenient for language processing tasks.

If you are using a computer with a CUDA-enabled graphics card, try to
install CUDA (>=8.0) and cuDNN (>=v6) and integrate them into TensorFlow.
Otherwise, you will still be able to complete the assignments in CPU mode
(but some of the models will be run on a smaller scale).

Some assignments will build on the tutorials and software available at

For convenience of download, it is highly recommended that you
have github installed on your computer (git is easy to install and
available for almost any operating system).



November 06, 2020 to December 04, 2020

Deep Learning


November 06, 2020 to December 04, 2020

Course Fee: $549

Do you meet course prerequisites? What about book & software? (Click here to learn more)

We have flexible policies to transfer to another course, or withdraw if necessary (modest fee applies)

Group rates: Email jdobbins "at" to get information on group rates. 

First time student or academic? Click here for an introductory offer on select courses. Academic affiliation?  You may be eligible for a discount at checkout.

Register Now

Add $50 service fee if you require a prior invoice, or if you need to submit a purchase order or voucher, pay by wire transfer or EFT, or refund and reprocess a prior payment. Please use this printed registration form, for these and other special orders.

Courses may fill up at any time and registrations are processed in the order in which they are received. Your registration will be confirmed for the first available course date, unless you specify otherwise.

The Institute for Statistics Education is certified to operate by the State Council of Higher Education in Virginia (SCHEV).

Contact Us
Have a question about a course before you register? Call us. We're here for you. (571) 281-8817 or ourcourses (at)

Want to be notified of future courses?

Student comments