World models: How AI learns to predict the consequences of actions
World models learn physical correlations through video and 3D simulation, and predict the consequences of their own actions, rather than learning language only from text. Thomas Brox, Professor of Pattern Recognition and Image Processing at the University of Freiburg, points out that this concept has long existed in biology and neuroscience, but is still fairly new in computer science; AlphaGo's internal representation was limited to the board, whereas current world models attempt to learn from data. Such models can allow self-driving cars and robots to rehearse everyday or dangerous scenarios in virtual environments, but the predictions are often difficult to understand and the technology is not yet mature.
World models learn physical correlations through video and 3D simulation, and predict the consequences of their own actions, rather than learning language only from text. Thomas Brox, Professor of Pattern Recognition and Image Processing at the University of Freiburg, points out that this concept has long existed in biology and neuroscience, but is still fairly new in computer science; AlphaGo's internal representation was limited to the board, whereas current world models attempt to learn from
Matched fieldsSummary、Body
