Get updates on Eligibility, Admission, Placements Fees Structure
Quick Facts
| Medium Of Instructions | Mode Of Learning | Mode Of Delivery |
|---|---|---|
| English | Self Study | Video and Text Based |
Course Overview
The University of Alberta and Alberta Machine Intelligence Institute offer the Sample-based Learning Methods training programme in conjunction with Coursera. An intermediate-level course, the programme is spread across four weeks, with intensive modules on reinforcement learning. The course will also help you understand how learning from experience can help you attain optimal behaviour.
The Sample-based Learning Methods certification course will help you understand concepts like Temporal Difference learning, Monte Carlo strategy, exploration, dynamic programming, and more. Moreover, the certification course will provide you with a chance to apply TD algorithms, Expected Sarsa, Q-learning, and more.
Instructors will conclude the Sample-based Learning Methods online course by looking at how algorithms can combine model-based planning and temporal difference updates to fast-track learning. Additionally, the certification course offers the flexibility to learn at your own pace in your own time and set your own deadlines.
You will also receive a combined shareable certificate from the University of Alberta and Coursera, upon completion of the Sample-based Learning Methods course.
The Highlights
- Course 2 of the 4-part Reinforcement Learning Specialisation
- Intermediate level
- Subtitles in Russian, English, Portuguese (Brazilian), French and Spanish
- Flexible deadlines
- Approx. 22 hour long course
- 100% online learning
- Shareable certificate
- Offered By University of Alberta and Alberta Machine Intelligence Institute
Programme Offerings
- Subtitles in Russian
- English
- Portuguese (Brazilian)
- French and Spanish
- Shareable Certificate
- Flexible Deadlines
- 22 hour long course
- 100% Online
- self paced learning options
- Course Readings
- practice quizzes
- Graded Assignments with peer feedback.
Courses and Certificate Fees
| Certificate Availability | Certificate Providing Authority |
|---|---|
| yes | University of Alberta, EdmontonCoursera |
Sample-based learning certification training program fee Structure :
Course Option | Fee |
1 Month (18 hours/week) | Rs. 6,757 |
3 Months (6 hours/week) | Rs. 13,514 |
6 Months (3 hours/week) | Rs. 20,271 |
Eligibility Criteria
Candidates need to know the fundamentals of linear algebra, probability, and Calculus to join the Sample-based Learning Methods online course certification by Coursera. Additionally, they need to have at least 1-year of working experience with Python 3.0 and should know how to use pseudocode to implement algorithms.
Moreover, candidates will get a course completion certificate after they successfully complete the Sample-based Learning Methods course.
What you will learn
Taking the Sample-based Learning Methods certification syllabus will help you:
- Learn machine learning, and reinforcement learning
- Apply the Temporal Difference algorithm for estimating value functions
- Comprehend the importance of exploration when employing sampled experience
- Understand two strategies for judging value functions from sampled experience- Temporal Difference and Monte Carlo
- Understand the relationship between Dynamic Programming and Temporal Difference and Monte Carlo
- Enforce a Dyna, a model-based approach to RL, which uses simulated experience
- Apply the two methods for control- Expected Sarsa and Q-learning
- Get an overview of planning and simulated experience along with artificial intelligence
- Differentiate between on-policy and off-policy control
- Carry out an empirical study to observe the improvements in sample efficiency while using Dyna
Who it is for
The course will be helping those candidates who would want to land up in any of the roles like
- AI Developer
- ML Engineer
- Data Scientist
- Data Engineer
- IT Engineer
Admission Details
You can seek admission to the Sample-based Learning Methods classes using the steps mentioned below:
- Log in to the official website of Coursera: www.coursera.org.
- Locate the “Sample-based Learning Methods” course page.
- Next, click on the “Enrol for free” tab on the bottom right side or top of the course page.
- You need to sign up using your email address or Google account to access the course.
- Choose the enrolment option based on your preference. You can choose from the paid mode or the ‘Full course, no certificate’ mode.
- Finally, pay the fee to complete the enrolment process.
Application Details
To apply for the Sample-based Learning Methods course by Coursera, you only need to sign up on Coursera. You can easily sign up using your Google account or email ID. In case you want to sign up using your email address, provide your name and create a password.
The Syllabus
Videos
- Meet your instructors
- Course Introduction
Readings
- Reinforcement Learning Textbook
- Read Me: Pre-requisites and Learning Objectives
Discussion Prompt
- Meet and Greet!
Videos
- What is Monte Carlo?
- Employing Monte Carlo for Action Values
- Importance of Sampling
- Employing Monte Carlo for Forecast
- Unravelling the Blackjack Example
- Employing Monte Carlo methods for non-specific policy iteration
- Why does off-policy learning matter?
- Epsilon-soft policies
- Emma Brunskill: Batch Reinforcement Learning
- Off-Policy Monte Carlo Forecast
- Week 1 Summary
Readings
- Module 1 Learning Objectives
- Weekly Reading
- Chapter Summary
Assignment
- Graded Quiz
Programming Assignment
- Blackjack
Discussion Prompt
- Comparing on-policy and off-policy learning
Videos
- What is Temporal Difference (TD) learning?
- Rich Sutton: The Importance of TD Learning
- The advantages of temporal difference learning
- Comparing TD and Monte Carlo
- Andy Barto and Rich Sutton: More on the History of RL
- Week 2 Summary
Readings
- Module 2 Learning Objectives
- Weekly Reading
Assignment
- Practice Quiz
Programming Assignment
- Policy Evaluation with Temporal Difference Learning
Discussion Prompt
- Should we care about TD in the brain?
Videos
- Sarsa: GPI with TD
- Sarsa in the Windy Grid World
- What is Q-learning?
- Q-learning in the Windy Grid World
- How is Q-learning off-policy?
- Expected Sarsa
- Expected Sarsa in the Cliff World
- Generality of Expected Sarsa
- Week 3 Summary
Readings
- Module 3 Learning Objectives
- Weekly Reading
- Chapter summary
Assignment
- Practice Quiz
Programming Assignment
- Q-Learning and Expected SARSA
Discussion Prompt
- How can we use off-policy for learning multiple goals?
Videos
- What is a Model?
- Comparing Sample and Distribution Models
- Random Tabular Q-planning
- The Dyna Architecture
- The Dyna Algorithm
- Dyna & Q-learning in a Simple Maze
- What if the model is inaccurate?
- In-depth with changing environments
- Drew Bagnell: self-driving, robotics, and Model Based RL
- Week 4 Summary
- Congratulations!
Readings
- Module 4 Learning Objectives
- Weekly Reading
- Chapter Summary
- Text Book Part 1 Summary
Assignments
- Practice Assessment
- Replacement Practice Assignment
Programming Assignment
- Dyna-Q and Dyna-Q+
Discussion Prompt
- Compare Planning and Reasoning