{"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/effective-approaches-to-batch-parallelization","title":"Effective Approaches to Batch Parallelization for Dynamic Neural Network Architectures","arxiv_id":"1707.02402","date":"2017-07-08","proceeding":null,"authors":["Joseph Suarez","Clare Zhu"],"abstract":"We present a simple dynamic batching approach applicable to a large class of\ndynamic architectures that consistently yields speedups of over 10x. We provide\nperformance bounds when the architecture is not known a priori and a stronger\nbound in the special case where the architecture is a predetermined balanced\ntree. We evaluate our approach on Johnson et al.'s recent visual question\nanswering (VQA) result of his CLEVR dataset by Inferring and Executing Programs\n(IEP). We also evaluate on sparsely gated mixture of experts layers and achieve\nspeedups of up to 1000x over the naive implementation.","url_abs":"http://arxiv.org/abs/1707.02402v1","url_pdf":"http://arxiv.org/pdf/1707.02402v1.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":"effective-approaches-to-batch-parallelization","repo_url":"https://github.com/jsuarez5341/Efficient-Dynamic-Batching","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"mixture-of-experts","task_name":"Mixture-of-Experts"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}