Emergence of Emotion Selectivity in Deep Neural Networks Trained to Recognize Visual Objects

<h1>Datasets and analysis code of the following publication:</h1> <p>Peng Liu, Ke Bo, Mingzhou Ding and Ruogu Fang (2024). Emergence of Emotion Selectivity in Deep Neural Networks &nbsp;Trained to Recognize Visual Objects.&nbsp;<em>PLOS Computational&nbsp;Biology.&nbsp;</em>DOI: 10.1371/journal.pcbi.1011943</p> <p>For any questions please contact the first author at mail pliu1 [at] ufl [dot] edu</p> <h2><strong>Contents:</strong></h2> <p><strong>&nbsp;Code_DataAnalysis</strong><br>&nbsp; - Extracted Selectivity</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; -- IAPS and NAPS datasets</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; -- Neurons In Alexnet and VGG networks</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; --Networks are pre-trained on ImageNet and randomly initialized</p> <p>&nbsp; - Extracted Overlapped Selectivity across IAPS and NAPS.</p> <p>&nbsp; - Extracted tuning performance changes from two datasets and the VGG network</p> <p>&nbsp; -Code to replicate the key results including&nbsp;</p> <p>&nbsp; &nbsp; &nbsp; &nbsp;--Tuning quality&nbsp;</p> <p>&nbsp; &nbsp; &nbsp; -- Number of overlapped neurons</p> <p>&nbsp; &nbsp; &nbsp; -- Enhance neuron activity</p> <p>&nbsp; &nbsp; &nbsp;-- Lesion neurons</p> <p>&nbsp;<strong>TrainedNetworks</strong></p> <p>&nbsp; &nbsp; &nbsp; &nbsp;--Pre-trained VGG network on ImageNet&nbsp;</p> <p>&nbsp; &nbsp; &nbsp; &nbsp;--Pre-trained Alexnet network on ImageNet</p> <p>After pre-training these networks on ImageNet, we fixed their weights and trained them to classify pleasant, neutral, and unpleasant images into three emotion categories using both IAPS and NAPS datasets.<br>&nbsp;</p> <p><strong>Image datasets</strong></p> <p>Access image datasets by request from https://csea.phhp.ufl.edu/media/iapsmessage.html for IAPS and https://lobi.nencki.edu.pl/research/8/ for NAPS.</p>