{"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/online-deep-learning-learning-deep-neural","title":"Online Deep Learning: Learning Deep Neural Networks on the Fly","arxiv_id":"1711.03705","date":"2017-11-10","proceeding":null,"authors":["Doyen Sahoo","Quang Pham","Jing Lu","Steven C. H. Hoi"],"abstract":"Deep Neural Networks (DNNs) are typically trained by backpropagation in a\nbatch learning setting, which requires the entire training data to be made\navailable prior to the learning task. This is not scalable for many real-world\nscenarios where new data arrives sequentially in a stream form. We aim to\naddress an open challenge of \"Online Deep Learning\" (ODL) for learning DNNs on\nthe fly in an online setting. Unlike traditional online learning that often\noptimizes some convex objective function with respect to a shallow model (e.g.,\na linear/kernel-based hypothesis), ODL is significantly more challenging since\nthe optimization of the DNN objective function is non-convex, and regular\nbackpropagation does not work well in practice, especially for online learning\nsettings. In this paper, we present a new online deep learning framework that\nattempts to tackle the challenges by learning DNN models of adaptive depth from\na sequence of training data in an online learning setting. In particular, we\npropose a novel Hedge Backpropagation (HBP) method for online updating the\nparameters of DNN effectively, and validate the efficacy of our method on\nlarge-scale data sets, including both stationary and concept drifting\nscenarios.","url_abs":"http://arxiv.org/abs/1711.03705v1","url_pdf":"http://arxiv.org/pdf/1711.03705v1.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":"online-deep-learning-learning-deep-neural","repo_url":"https://github.com/LIBOL/ODL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"online-deep-learning-learning-deep-neural","repo_url":"https://github.com/Rohit102497/Aux-Drop","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"online-deep-learning-learning-deep-neural","repo_url":"https://github.com/alison-carrera/onn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"online-deep-learning-learning-deep-neural","repo_url":"https://github.com/meyresearch/online_vampnets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"online-deep-learning-learning-deep-neural","repo_url":"https://github.com/phquang/OnlineDeepLearning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[{"method_slug":"odl","method_name":"ODL"}],"datasets_introduced":[],"methods_introduced":[{"slug":"odl","name":"ODL","full_name":"online deep learning"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.03705","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}