An AI learns physics concepts like a child discovering the world


A new study is inspired by the theory of developmental psychology to teach an artificial intelligence some fundamentals of physics. Shared in the journal Nature, the study has been carried out by a team of researchers from the company DeepMind. They created a program that can learn simple physical rules about how objects behave Learning ‘common sense’ While AI has made amazing progress in recent years, the most advanced systems still struggle to exercise ‘common sense’ . Common sense that guides prediction, inference, and action in everyday scenarios. DeepMind scientists admit, however, that this intuitive physics is “fundamental to embodied intelligence,” because “it is essential to any practical interaction,” they write. To carry out their test, the research team led by Luis Piloto was inspired by psychology. of child development. Developmental psychologists study how babies follow the movement of objects by following their gaze. When shown a video of a ball suddenly disappearing, for example, young children express surprise. Thus, researchers measure their surprise by the duration of their gaze in a certain direction, this is precisely what the DeepMind team of scientists wanted to discover, training an AI called “PLATO” with videos of simple objects, like balls rolling and crashing. solidity, immutability and directional inertia. “Each video with a violation of these physics principles is associated with a corresponding video that provides a baseline that conforms to the physics principles, maintaining precisely the same single-frame statistics for all videos,” the researchers explain. The dataset also includes a separate corpus of videos intended to serve as training data. he calls “intuitive physics” like a child discovering the world. The method used to measure knowledge of a physical concept is called the “expectancy violation paradigm,” the researchers say. 28 hours of videos The researchers observed learning effects after just 28 hours of videos. Some progress, even if they acknowledge that “the range of object types and events in our dataset remains narrow compared to those found in the real world,” he asked about the potential implications of his work for to developmental psychology itself. “This topic must be approached with some caution, since the model we presented does not aim to provide a direct model of the acquisition of physical concepts in children. However, we believe that there are several insights that may be useful for developmental science. First, our modeling work provides a proof-of-concept showing that at least some central concepts of intuitive physics can be learned through visual learning. Advertisement Your content continues below
#learns #physics #concepts #child #discovering #world


Please enter your comment!
Please enter your name here