{"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/using-stochastic-computation-graphs-formalism","title":"Using stochastic computation graphs formalism for optimization of sequence-to-sequence model","arxiv_id":"1711.07724","date":"2017-11-21","proceeding":null,"authors":["Eugene Golikov","Vlad Zhukov","Maksim Kretov"],"abstract":"Variety of machine learning problems can be formulated as an optimization\ntask for some (surrogate) loss function. Calculation of loss function can be\nviewed in terms of stochastic computation graphs (SCG). We use this formalism\nto analyze a problem of optimization of famous sequence-to-sequence model with\nattention and propose reformulation of the task. Examples are given for machine\ntranslation (MT). Our work provides a unified view on different optimization\napproaches for sequence-to-sequence models and could help researchers in\ndeveloping new network architectures with embedded stochastic nodes.","url_abs":"http://arxiv.org/abs/1711.07724v2","url_pdf":"http://arxiv.org/pdf/1711.07724v2.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":"using-stochastic-computation-graphs-formalism","repo_url":"https://github.com/deepmipt/seq2seq_scg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}