Researchers used deep learning to analyze brain cells by exposing the algorithm to both stained and unstained images. The program was able to identify where cells were located in culture dishes by learning to spot the cell's nucleus.
Researchers have created a deep learning algorithm that can be tricked by optical illusions, just like humans. The findings shed new light on the human visual system and may help improve artificial vision.
Using ECoG and machine learning, researchers decoded spoken words and phrases in real-time from brain signals that control speech. The technology could eventually be used to help those who have lost vocal control to regain their voice.
Using a Siamese neural network, a new deep learning system dubbed Ghostwriter is able to distinguish the writing styles between different texts. The algorithm compares a student's current written work against past examples of their prose, providing a percentage score for similar writing styles. The new system could be used to determine if a student has plagiarized their work, or if it is an original piece.
Examining the cognitive abilities of the AI language model, GPT-3, researchers found the algorithm can keep up and compete with humans in some areas but falls behind in others due to a lack of real-world experience and interactions.
The brain tunes out information from the outside world, such as the sound of speech, during REM sleep. During light sleep, the brain prioritizes meaningful speech, just as it does during a wakeful state. Researchers believe the mechanism enables the brain to protect the dreaming phase, which is essential for memory consolidation.
Applying AI technology, researchers teach a robot to make appropriate reactive human facial expressions. Scientists say this could help build trust between humans and the humanoid robots they interact with.