{"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/the-streaming-rollout-of-deep-networks","title":"The streaming rollout of deep networks - towards fully model-parallel execution","arxiv_id":"1806.04965","date":"2018-06-13","proceeding":"NeurIPS 2018 12","authors":["Volker Fischer","Jan Köhler","Thomas Pfeil"],"abstract":"Deep neural networks, and in particular recurrent networks, are promising\ncandidates to control autonomous agents that interact in real-time with the\nphysical world. However, this requires a seamless integration of temporal\nfeatures into the network's architecture. For the training of and inference\nwith recurrent neural networks, they are usually rolled out over time, and\ndifferent rollouts exist. Conventionally during inference, the layers of a\nnetwork are computed in a sequential manner resulting in sparse temporal\nintegration of information and long response times. In this study, we present a\ntheoretical framework to describe rollouts, the level of model-parallelization\nthey induce, and demonstrate differences in solving specific tasks. We prove\nthat certain rollouts, also for networks with only skip and no recurrent\nconnections, enable earlier and more frequent responses, and show empirically\nthat these early responses have better performance. The streaming rollout\nmaximizes these properties and enables a fully parallel execution of the\nnetwork reducing runtime on massively parallel devices. Finally, we provide an\nopen-source toolbox to design, train, evaluate, and interact with streaming\nrollouts.","url_abs":"http://arxiv.org/abs/1806.04965v2","url_pdf":"http://arxiv.org/pdf/1806.04965v2.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":"the-streaming-rollout-of-deep-networks","repo_url":"https://github.com/boschresearch/statestream","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.04965","atlas_url":"https://app.syntology.ai/?focus=1806.04965","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1806.04965"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/boschresearch/statestream","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":10},"by_repo_kind":{"official":{"samples":10,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"cb52df5a86515237","entry":"S2L","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/meta/network.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/meta/network.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cb52df5a86515237"}},{"code_sha256_prefix":"43a151508d379182","entry":"cm2fd","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/meta/network.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/meta/network.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"43a151508d379182"}},{"code_sha256_prefix":"a8c1507b69fb2a97","entry":"generate_graph","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/utils/wrapper_networkx.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/utils/wrapper_networkx.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a8c1507b69fb2a97"}},{"code_sha256_prefix":"62c65ea4bfe94610","entry":"get_item_type","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/meta/network.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/meta/network.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"62c65ea4bfe94610"}},{"code_sha256_prefix":"daebb218663a3f81","entry":"get_value","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/backends/backend_tensorflow.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/backends/backend_tensorflow.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"daebb218663a3f81"}},{"code_sha256_prefix":"87df0e2f54903130","entry":"get_value","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/backends/backend_theano.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/backends/backend_theano.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"87df0e2f54903130"}},{"code_sha256_prefix":"ecf1b93e05907804","entry":"has_target","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/meta/losses.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/meta/losses.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ecf1b93e05907804"}},{"code_sha256_prefix":"50eda2dd66d22b38","entry":"import_backend","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/backends/backends.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/backends/backends.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"50eda2dd66d22b38"}},{"code_sha256_prefix":"e94f3479871d150e","entry":"scalar","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/backends/backend_tensorflow.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/backends/backend_tensorflow.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e94f3479871d150e"}},{"code_sha256_prefix":"c0eee06fb792d65d","entry":"variable","repo":"boschresearch/statestream","repo_kind":"official","path":"statestream/backends/backend_tensorflow.py","file_url":"https://github.com/boschresearch/statestream/blob/HEAD/statestream/backends/backend_tensorflow.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c0eee06fb792d65d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}