Calvin

@techreport{arxiv2018deepca,
title = {Deep Component Analysis via Alternating Direction Neural Networks},
author = {Calvin Murdock and Ming-Fang Chang and Simon Lucey},
url = {https://arxiv.org/pdf/1803.06407.pdf, Paper},
year = {2018},
date = {2018-03-19},
institution = {Carnegie Mellon University},
abstract = {Despite a lack of theoretical understanding, deep neural networks have achieved unparalleled performance in a wide range of applications. On the other hand, shallow representation learning with component analysis is associated with rich intuition and theory, but smaller capacity often limits its usefulness. To bridge this gap, we introduce Deep Component Analysis (DeepCA), an expressive multilayer model formulation that enforces hierarchical structure through constraints on latent variables in each layer. For inference, we propose a differentiable optimization algorithm implemented using recurrent Alternating Direction Neural Networks (ADNNs) that enable parameter learning using standard backpropagation. By interpreting feed-forward networks as single-iteration approximations of inference in our model, we provide both a novel theoretical perspective for understanding them and a practical technique for constraining predictions with prior knowledge. Experimentally, we demonstrate performance improvements on a variety of tasks, including single-image depth prediction with sparse output constraints.},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}

@inproceedings{murdock2017grass,
title = {Approximate Grassmannian Intersections: Subspace-Valued Subspace Learning},
author = {Calvin Murdock and Fernando {De la Torre}},
url = {http://www.calvinmurdock.com/content/uploads/publications/iccv2017grass.pdf, Paper
http://www.calvinmurdock.com/agi/, Project Page},
year = {2017},
date = {2017-10-22},
booktitle = {International Conference on Computer Vision (ICCV)},
abstract = {Subspace learning is one of the most foundational tasks in computer vision with applications ranging from dimensionality reduction to data denoising. As geometric objects, subspaces have also been successfully used for efficiently representing certain types of invariant data. However, methods for subspace learning from subspace-valued data have been notably absent due to incompatibilities with standard problem formulations. To fill this void, we introduce Approximate Grassmannian Intersections (AGI), a novel geometric interpretation of subspace learning posed as finding the approximate intersection of constraint sets on a Grassmann manifold. Our approach can naturally be applied to input subspaces of varying dimension while reducing to standard subspace learning in the case of vector-valued data. Despite the nonconvexity of our problem, its globally-optimal solution can be found using a singular value decomposition. Furthermore, we also propose an efficient, general optimization approach that can incorporate additional constraints to encourage properties such as robustness. Alongside standard subspace applications, AGI also enables the novel task of transfer learning via subspace completion. We evaluate our approach on a variety of applications, demonstrating improved invariance and generalization over vector-valued alternatives.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}

@inproceedings{murdock2016blockout,
title = {Blockout: Dynamic Model Selection for Hierarchical Deep Networks},
author = {Calvin Murdock and Zhen Li and Howard Zhou and Tom Duerig},
url = {http://www.calvinmurdock.com/content/uploads/publications/cvpr2016blockout.pdf, Paper
http://arxiv.org/pdf/1512.05246v1.pdf, Extended Technical Report
http://www.calvinmurdock.com/blockout/, Project Page},
year = {2016},
date = {2016-06-27},
booktitle = {Conference on Computer Vision and Pattern Recognition (CVPR)},
abstract = {Most deep architectures for image classification--even those that are trained to classify a large number of diverse categories--learn shared image representations with a single model. Intuitively, however, categories that are more similar should share more information than those that are very different. While hierarchical deep networks address this problem by learning separate features for subsets of related categories, current implementations require simplified models using fixed architectures specified via heuristic clustering methods. Instead, we propose Blockout, a method for regularization and model selection that simultaneously learns both the model architecture and parameters. A generalization of Dropout, our approach gives a novel parametrization of hierarchical architectures that allows for structure learning via back-propagation. To demonstrate its utility, we evaluate Blockout on the CIFAR and ImageNet datasets, demonstrating improved classification accuracy, better regularization performance, faster training, and the clear emergence of hierarchical network structures.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}