Fig. 2: Representational embeddings inferred from human and DNN behaviour.
From: Dimensions underlying the representational alignment of deep neural networks with humans
a, Visualization of example dimensions from human- and DNN-derived representational embeddings, with a selection of dimensions that had been rated as semantic, mixed visual–semantic and visual, alongside their dimension labels obtained from human judgements. Note that the displayed images reflect only images with a public domain license and not the full image set76. b, Rating procedure for each dimension, which was based on visualizing the top k images according to their numeric weights. Human participants labelled each of the human and DNN dimensions as predominantly semantic, visual, mixed visual–semantic or unclear (unclear ratings are not shown; 7.35% of all dimensions are for humans and 8.57%, for VGG-16). c, Relative importance of dimensions labelled as visual and semantic, where VGG-16 exhibited a dominance of visual and mixed dimensions relative to humans that showed a clear dominance of semantic dimensions. Images in a and b reproduced with permission from ref. 76, Springer Nature Limited.
