Relating sentence representations in deep neural networks with those encoded by the brain

Researchers at the Indian Institute of Science (IISc) and Carnegie Mellon University (CMU) have recently carried out a study exploring the relationship between sentence representations acquired by deep neural networks and those encoded by the brain. Their paper, pre-published on arXiv and set to be presented at this year's Association for Computational Linguistics (ACL) conference, unveiled correlations between activations in deep neural models and MEG brain data that could aid our current understanding of how the brain and deep learning algorithms process language.

source https://www.lifetechnology.com/blogs/life-technology-news-blog/relating-sentence-representations-in-deep-neural-networks-with-those-encoded-by-the-brain