Chinese researchers create AI to understand animal social behavior

Understanding the social interactions of animals holds significant importance for studying their behavior and its implications in neuroscience and ecology. Researchers at the Brain Cognition and Brain Disease Institute (BCBDI) of the Shenzhen Institute of Advanced Technology of the Chinese Academy of Sciences have introduced a pioneering method known as the Social Behavior Atlas (SBeA). This innovative approach utilizes artificial intelligence (AI) to track and analyze the behavior of multiple animals in three dimensions (3D).

Published in Nature Machine Intelligence, this study presents a groundbreaking technique for analyzing multi-animal behavior. Unlike conventional methods that rely on predefined categories of social behavior, the SBeA framework employs a few-shot learning AI algorithm. This algorithm can accurately identify visually similar animals with over 90 percent accuracy, enabling the discovery of previously undefined differences in animal social behavior.

A key advantage of the SBeA framework is its capacity to synthesize vast amounts of data and train models with heightened accuracy. According to Wei Pengfei, the corresponding author of the study, this leads to more precise estimation of 3D social gestures.

The SBeA technology proves particularly effective in determining 3D social posture, recognizing individual animals, and analyzing subtle social interactions across various species such as mice, birds, and domestic dogs. Moreover, its potential for cross-species applications opens up new avenues for studying social behavior across different animal types.

Leave a Reply

Your email address will not be published.