World Models and Autonomous Machine Intelligence

Abstract

The PI, Dr. Yann LeCun, is on a ten year quest based on work towards his Joint Embedding Predictive Architecture (JEPA) system. This work makes use of many bodies of prior work such as cognitive science, systems neuroscience, optimal control theory, reinforcement learning, traditional AI, and integrates them with new concepts in machine learning such as self-supervised learning and joint-embedding architectures. The architecture of JEPA comprises diverse differentiable modules that compute gradient estimates objective functions that can propagate “information� to other “upstream� modules. The outcome being sets of modules and their inter-connections to interwork the “knowledge� they compose via their inter-connections and the function to reason and predict the value-truth in some new learned “thing�; the knowledge learned.

Document Details

• Document Type: DoD Grant Award
• Publication Date: Feb 29, 2024
• Source ID: FA95502310139

Entities

People

• Yann Le Cun

Organizations

• Air Force Office of Scientific Research
• New York University
• United States Air Force

Tags

Fields of Study

• Computer science

Readers

• Neural Network Machine Learning.
• Systems Analysis and Design

Technology Areas

• AI & ML
• AI & ML - DoD AI Strategy
• AI & ML - Machine Learning Algorithms
• AI & ML - Neural Networks