AI network simulations become unstable following continuous periods of unsupervised learning. When the networks were exposed to states that are analogous to human brain waves during sleep, the stability was restored.
A new deep learning algorithm helped researchers identify a powerful new antibiotic compound that kills many of the world's most problematic, disease-causing bacterias, including those which have so far been resistant to common antibiotics.
EmoNet, a new convolutional neural network, can accurately decode images into eleven distinct emotional categories. Training the AI on over 25,000 images, researchers demonstrate image content is sufficient to predict the category and valence of human emotions.