Understanding Artificial Intelligence through Algorithmic Information Theory

Overview

Delve into the depths of Artificial Intelligence (AI) with the groundbreaking perspectives offered by Algorithmic Information Theory. This innovatively designed course unveils the essence of AI, beyond its complex problem-solving facade. Whether you're venturing into the realms of machine learning, keen on the nuances of reasoning, fascinated by the mathematical underpinnings, or curious about the parallels with human intelligence, this course serves as a critical roadmap.

Discover how AI transcends traditional boundaries, with insights into the formulation of optimal hypotheses for learning tasks, the intricacies of learning techniques like clustering or neural networks viewed through the lens of information compression. Explore how reasoning, in all its forms, essentially condenses complex descriptions, and how probability and randomness acquire new definitions through algorithmic information, shedding light on the theoretical constraints of AI.

Unravel the contributions of three pioneering mathematicians who, half a century ago, established the profound connection between information and computer science, giving birth to Algorithmic Information Theory. This theory, central to the theoretical foundations of AI, champions the elimination of redundancy to distill meaningful information, thereby guiding AI towards making optimal decisions and delineating its limitations.

Designed by Institut Mines-Télécom and provided by edX, this course falls under the categories of Artificial Intelligence Courses, Machine Learning Courses, and Algorithms and Data Structures Courses. Imbued with essential keywords such as Algorithmic information, Kolmogorov complexity, theoretical computer science, and more, it promises a comprehensive exploration of AI's theoretical and practical dimensions.

Syllabus

Taught by

Jean-Louis Dessalles

Tags

france