{"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-recurrent-gaussian-process-with","title":"Deep Recurrent Gaussian Process with Variational Sparse Spectrum Approximation","arxiv_id":"1711.00799","date":"2017-11-02","proceeding":null,"authors":["Roman Föll","Bernard Haasdonk","Markus Hanselmann","Holger Ulmer"],"abstract":"Modeling sequential data has become more and more important in practice. Some\napplications are autonomous driving, virtual sensors and weather forecasting.\nTo model such systems so called recurrent models are used. In this article we\nintroduce two new Deep Recurrent Gaussian Process (DRGP) models based on the\nSparse Spectrum Gaussian Process (SSGP) and the improved variational version\ncalled Variational Sparse Spectrum Gaussian Process (VSSGP). We follow the\nrecurrent structure given by an existing DRGP based on a specific sparse\nNystr\\\"om approximation. Therefore, we also variationally integrate out the\ninput-space and hence can propagate uncertainty through the layers. We can show\nthat for the resulting lower bound an optimal variational distribution exists.\nTraining is realized through optimizing the variational lower bound. Using\nDistributed Variational Inference (DVI), we can reduce the computational\ncomplexity. We improve over current state of the art methods in prediction\naccuracy for experimental data-sets used for their evaluation and introduce a\nnew data-set for engine control, named Emission. Furthermore, our method can\neasily be adapted for unsupervised learning, e.g. the latent variable model and\nits deep version.","url_abs":"http://arxiv.org/abs/1711.00799v2","url_pdf":"http://arxiv.org/pdf/1711.00799v2.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-recurrent-gaussian-process-with","repo_url":"https://github.com/RomanFoell/DRGP-VSS","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"variational-inference","task_name":"Variational Inference"},{"task_slug":"weather-forecasting","task_name":"Weather Forecasting"}],"methods":[{"method_slug":"gaussian-process","method_name":"Gaussian Process"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}