A newborn embodied Turing test for view-invariant object recognition
This work addresses the challenge of developing AI that learns like animals, though it is incremental as it primarily establishes a comparative framework rather than achieving parity.
The researchers tackled the problem of comparing learning abilities between newborn animals and AI by creating a 'newborn embodied Turing Test' where both were raised in identical environments and tested on the same tasks, finding that machines could mimic imprinting behavior but performed far worse on view-invariant object recognition, with chicks achieving near-universal success while machines mostly failed.
Recent progress in artificial intelligence has renewed interest in building machines that learn like animals. Almost all of the work comparing learning across biological and artificial systems comes from studies where animals and machines received different training data, obscuring whether differences between animals and machines emerged from differences in learning mechanisms versus training data. We present an experimental approach-a "newborn embodied Turing Test"-that allows newborn animals and machines to be raised in the same environments and tested with the same tasks, permitting direct comparison of their learning abilities. To make this platform, we first collected controlled-rearing data from newborn chicks, then performed "digital twin" experiments in which machines were raised in virtual environments that mimicked the rearing conditions of the chicks. We found that (1) machines (deep reinforcement learning agents with intrinsic motivation) can spontaneously develop visually guided preference behavior, akin to imprinting in newborn chicks, and (2) machines are still far from newborn-level performance on object recognition tasks. Almost all of the chicks developed view-invariant object recognition, whereas the machines tended to develop view-dependent recognition. The learning outcomes were also far more constrained in the chicks versus machines. Ultimately, we anticipate that this approach will help researchers develop embodied AI systems that learn like newborn animals.