Shewmake, C., Buracas, D., Lillemark, H., Shin, J., Bekkers, E., Miolane, N., Olshausen, B.
Abstract
We propose a hierarchical neural network architecture for unsupervised learning of equiv-ariant part-whole decompositions of visual scenes. In contrast to the global equivariance of group-equivariant networks, the proposed architecture exhibits equivariance to part-whole transformations throughout the hierarchy, which we term hierarchical equivariance. The model achieves these structured internal representations via hierarchical Bayesian inference, which gives rise to rich bottom-up, top-down, and lateral information flows, hypothesized to underlie the mechanisms of perceptual inference in visual cortex. We demonstrate these useful properties of the model on a simple dataset of scenes with multiple objects under independent rotations and translations.
Citation
Shewmake, C., Buracas, D., Lillemark, H., Shin, J., Bekkers, E., Miolane, N., & Olshausen, B. (2023, December). Visual Scene Representation with Hierarchical Equivariant Sparse Coding. Proceedings of Machine Learning Research-NeurIPS workshop.
BibTeX
@article{osti_10529135, place = {Country unknown/Code not available}, title = {Visual Scene Representation with Hierarchical Equivariant Sparse Coding}, url = {https://par.nsf.gov/biblio/10529135}, journal = {}, publisher = {Proceedings of Machine Learning Research - NeurIPS workshop}, author = {Shewmake, Christian and Buracas, Domas and Lillemark, Hansen and Shin, Jinho and Bekkers, Erik and Miolane, Nina and Olshausen, Bruno}, }
