{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deep-component-analysis-via-alternating","title":"Deep Component Analysis via Alternating Direction Neural Networks","arxiv_id":"1803.06407","date":"2018-03-16","proceeding":"ECCV 2018 9","authors":["Calvin Murdock","Ming-Fang Chang","Simon Lucey"],"abstract":"Despite a lack of theoretical understanding, deep neural networks have\nachieved unparalleled performance in a wide range of applications. On the other\nhand, shallow representation learning with component analysis is associated\nwith rich intuition and theory, but smaller capacity often limits its\nusefulness. To bridge this gap, we introduce Deep Component Analysis (DeepCA),\nan expressive multilayer model formulation that enforces hierarchical structure\nthrough constraints on latent variables in each layer. For inference, we\npropose a differentiable optimization algorithm implemented using recurrent\nAlternating Direction Neural Networks (ADNNs) that enable parameter learning\nusing standard backpropagation. By interpreting feed-forward networks as\nsingle-iteration approximations of inference in our model, we provide both a\nnovel theoretical perspective for understanding them and a practical technique\nfor constraining predictions with prior knowledge. Experimentally, we\ndemonstrate performance improvements on a variety of tasks, including\nsingle-image depth prediction with sparse output constraints.","url_abs":"http://arxiv.org/abs/1803.06407v1","url_pdf":"http://arxiv.org/pdf/1803.06407v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deep-component-analysis-via-alternating","repo_url":"https://github.com/DeadAt0m/DCA-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"depth-prediction","task_name":"Depth Prediction"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}