Convolutional neural networks can decode whether a person is listening to a speaker on the left or right solely from 1 to 2s of EEG data.

Convolutional neural networks are an effective tool for detecting archaic adaptive introgression from genomic data.

Mirror-symmetric view tuning in the macaque AL face patch can be explained by the spatial pooling of learned reflection-equivariant representations.

A novel deep-learning framework shows how to interpret and decode raw neural recordings, avoiding the need for strong prior hypotheses, revealing a novel representation of head direction.

A new method is described to identify sharp-wave ripples from the rodent hippocampus with deep learning techniques, which may help to identify and characterize previously undetected physiological events.

Scenario-invariant representation of physical stability is found in the frontoparietal regions but not the ventral visual pathway (and ImageNet-trained convolutional neural networks), consistent with the hypothesis that computations underlying physical stability may entail forward simulations of what will happen next.

Machine learning has been applied to develop predictive models for designing RNA gene-control elements with targeted gene-regulatory activities.

Implementing neural changes associated with attention in a deep neural network causes performance changes that mimic those observed in humans and macaques.

Convolutional neural networks and graph partitioning algorithms can be combined into an easy-to-use tool for segmentation of cells in dense plant tissue volumes imaged with light microscopy.

The animacy organization in human ventral visual cortex is driven by both the presence of animal-diagnostic visual features and the psychological property of agency.