Variational autoencoder: An unsupervised model for encoding and decoding fMRI activity in visual cortex

• Published in: NeuroImage (Orlando, Fla.) Vol. 198; pp. 125 - 136
• Main Authors: Han, Kuan, Wen, Haiguang, Shi, Junxing, Lu, Kun-Han, Zhang, Yizhen, Fu, Di, Liu, Zhongming
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.09.2019; Elsevier Limited
• Subjects: Adult; Artificial intelligence; Bayesian brain; Brain Mapping - methods; Computer applications; Deep learning; Economic models; Female; Functional magnetic resonance imaging; Humans; Hypotheses; Image Processing, Computer-Assisted; Magnetic Resonance Imaging; Models, Neurological; Neural coding; Neural networks; Neural Networks, Computer; Neuroimaging; Neurosciences; Pattern Recognition, Visual - physiology; Random variables; Supervision; Unsupervised Machine Learning; Variational autoencoder; Visual cortex; Visual Cortex - physiology; Visual discrimination learning; Visual reconstruction; Young Adult; Neural coding; Variational autoencoder; Bayesian brain; Visual reconstruction

Goal-driven and feedforward-only convolutional neural networks (CNN) have been shown to be able to predict and decode cortical responses to natural images or videos. Here, we explored an alternative deep neural network, variational auto-encoder (VAE), as a computational model of the visual cortex. W...