{"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/forward-thinking-building-and-training-neural","title":"Forward Thinking: Building and Training Neural Networks One Layer at a Time","arxiv_id":"1706.02480","date":"2017-06-08","proceeding":null,"authors":["Chris Hettinger","Tanner Christensen","Ben Ehlert","Jeffrey Humpherys","Tyler Jarvis","Sean Wade"],"abstract":"We present a general framework for training deep neural networks without\nbackpropagation. This substantially decreases training time and also allows for\nconstruction of deep networks with many sorts of learners, including networks\nwhose layers are defined by functions that are not easily differentiated, like\ndecision trees. The main idea is that layers can be trained one at a time, and\nonce they are trained, the input data are mapped forward through the layer to\ncreate a new learning problem. The process is repeated, transforming the data\nthrough multiple layers, one at a time, rendering a new data set, which is\nexpected to be better behaved, and on which a final output layer can achieve\ngood performance. We call this forward thinking and demonstrate a proof of\nconcept by achieving state-of-the-art accuracy on the MNIST dataset for\nconvolutional neural networks. We also provide a general mathematical\nformulation of forward thinking that allows for other types of deep learning\nproblems to be considered.","url_abs":"http://arxiv.org/abs/1706.02480v1","url_pdf":"http://arxiv.org/pdf/1706.02480v1.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":"forward-thinking-building-and-training-neural","repo_url":"https://github.com/tkchris93/ForwardThinking","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.02480","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}