Reading List #1

As noted in the previous post, I'm going to start using this blog as a public reading list, or a "recently read and interesting" list. Here are the first 3 that I read recently and liked.

Yamins, D. L. K., Hong, H., Cadieu, C. F., Solomon, E. A., Seibert, D., & DiCarlo, J. J. (2014). Performance-optimized hierarchical models predict neural responses in higher visual cortex. Proceedings of the National Academy of Sciences of the United States of America, 111(23), 8619–24. http://doi.org/10.1073/pnas.1403112111

Kümmerer, M., Theis, L., & Bethge, M. (2014). Deep Gaze I-Boosting Saliency Prediction with Feature Maps Trained on ImageNet. arXiv:1411.1045 [Cs, Q-Bio, Stat], (2014). Retrieved from http://arxiv.org/abs/1411.1045%5Cnfiles/1004/arXiv-Kummerer_et_al-2014-Deep_Gaze_I-Boosting_Saliency_Prediction_with_Feature_Maps_Trained_on_ImageNet.pdf

Hong, H., Yamins, D. L. K., Majaj, N. J., & DiCarlo, J. J. (2016). Explicit information for category-orthogonal object properties increases along the ventral stream. Nature Neuroscience, 19(4), 613–622.http://doi.org/10.1038/nn.4247

Yamins et al. (2014) and Hong et al. (2016) are both from the same group that I first learned about during the CBMM (Center for Brains Minds and Machines) course. I was quite surprised by the results in the 2014 paper and it definitely affected the direction of my research. 

Kummerer et al. (2014) basically takes the idea from Yamins et al. (2014) and applies it to my research area. He uses a set of features learned in image classification and trains a simple model to weigh those features to predict fixation maps.