Course

Reinforcement Learning

Alberta Machine Intelligence Institute & University of Alberta

The Reinforcement Learning Specialization consists of 4 comprehensive courses that delve into the power of adaptive learning systems and artificial intelligence (AI). Throughout the specialization, you will learn how to build a Reinforcement Learning system for sequential decision making and gain an understanding of the space of RL algorithms, including Temporal-Difference learning, Monte Carlo, Sarsa, Q-learning, Policy Gradients, Dyna, and more. You will also grasp how to formalize tasks as Reinforcement Learning problems and comprehend how RL fits within the broader umbrella of machine learning, complementing deep learning, supervised, and unsupervised learning.

This specialization is designed to equip learners with the skills to apply RL tools to various domains such as game development, customer interaction, smart assistants, recommender systems, supply chain, industrial control, finance, and more. By the end of the specialization, you will have the knowledge and practical experience to understand the foundations of modern probabilistic AI and apply AI tools and ideas to real-world problems.

Build a Reinforcement Learning system for sequential decision making
Understand the space of RL algorithms
Formalize tasks as Reinforcement Learning problems
Comprehend how RL fits within the broader umbrella of machine learning

Certificate Available ✔

Get Started / More Info

Gain in-depth knowledge of Reinforcement Learning through 4 courses covering the fundamentals, sample-based learning methods, prediction and control with function approximation, and a complete RL solution capstone project.

Fundamentals of Reinforcement Learning

Formalize problems as Markov Decision Processes and understand basic exploration methods and the exploration / exploitation tradeoff. You will also learn about value functions and dynamic programming as an efficient solution approach to industrial control problems.

Sample-based Learning Methods

Learn about algorithms that can learn near optimal policies based on trial and error interaction with the environment. Understand Temporal-Difference learning, Monte Carlo, and how to get the best of both worlds with algorithms that combine model-based planning and temporal difference updates.

Prediction and Control with Function Approximation

Explore how to solve problems with large, high-dimensional, and potentially infinite state spaces. Understand how to use supervised learning approaches to approximate value functions, implement TD with function approximation, and apply policy gradient methods in a continuous-action environment.

A Complete Reinforcement Learning System (Capstone)

Implement a complete RL solution to a problem, starting from problem formulation, appropriate algorithm selection and implementation, and an empirical study into the effectiveness of the solution. This capstone project lets you see how each component fits together into a complete solution.

Course

Reinforcement Learning

Course Modules

Fundamentals of Reinforcement Learning

Sample-based Learning Methods

Prediction and Control with Function Approximation

A Complete Reinforcement Learning System (Capstone)

More Machine Learning Courses

Despliegue de modelos Deep Learning: Flask, heroku, Postman

Linear Regression with NumPy and Python

Simple Nearest Neighbors Regression and Classification

Github CoPilot for Beginners: Write Software With AI