Numerical methods (NM) - KAIN6M05

Goal(s)

Numerical methods are a set of mathematical techniques used to solve numerical problems, often when exact analytical solutions are not available. They are widely used in engineering, physics, computer science, and many other scientific fields.

This course demonstrates the use of numerical methods in Artificial Intelligence by presenting the theoretical and practical foundations of reinforcement learning. This technique is based on mathematical models such as Markov Decision Processes (MDPs), Bellman equations, and optimization methods.

The application of numerical methods in reinforcement learning involves solving equations (e.g., successive iterations in Q-Learning or gradient-based methods used in Policy Gradient) and using optimization approaches (e.g., algorithms like Deep Q-Learning or Actor-Critic).

Content(s)

Chapter 1: Introduction
Chapter 2: Mathematical Concepts
Chapter 3: Reinforcement Learning
Chapter 4: Multi-Armed Bandit Problem
Chapter 5: Markov Decision Processes
Chapter 6: Dynamic Programming and RL Algorithms

L2 level in Mathematics (analysis and algebra), programming and algorithms

Calendar

The course exists in the following branches:

• Curriculum - INFO - Semester 6

Additional Information

Course ID : KAIN6M05
Course language(s):

You can find this course among all other courses.