{"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/jointly-learned-exit-and-inference-for-a","title":"Jointly-Learned Exit and Inference for a Dynamic Neural Network : JEI-DNN","arxiv_id":"2310.09163","date":"2023-10-13","proceeding":null,"authors":["Florence Regol","Joud Chataoui","Mark Coates"],"abstract":"Large pretrained models, coupled with fine-tuning, are slowly becoming established as the dominant architecture in machine learning. Even though these models offer impressive performance, their practical application is often limited by the prohibitive amount of resources required for every inference. Early-exiting dynamic neural networks (EDNN) circumvent this issue by allowing a model to make some of its predictions from intermediate layers (i.e., early-exit). Training an EDNN architecture is challenging as it consists of two intertwined components: the gating mechanism (GM) that controls early-exiting decisions and the intermediate inference modules (IMs) that perform inference from intermediate representations. As a result, most existing approaches rely on thresholding confidence metrics for the gating mechanism and strive to improve the underlying backbone network and the inference modules. Although successful, this approach has two fundamental shortcomings: 1) the GMs and the IMs are decoupled during training, leading to a train-test mismatch; and 2) the thresholding gating mechanism introduces a positive bias into the predictive probabilities, making it difficult to readily extract uncertainty information. We propose a novel architecture that connects these two modules. This leads to significant performance improvements on classification datasets and enables better uncertainty characterization capabilities.","url_abs":"https://arxiv.org/abs/2310.09163v2","url_pdf":"https://arxiv.org/pdf/2310.09163v2.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":"jointly-learned-exit-and-inference-for-a","repo_url":"https://github.com/networkslab/dynn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"dynamic-neural-networks","task_name":"Dynamic neural networks"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2310.09163","atlas_url":"https://app.syntology.ai/?focus=2310.09163","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.09163"}},"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/networkslab/dynn","reach":{"status":"ok"}}],"summary":{"ran":9},"by_repo_kind":{"official":{"samples":9,"ran":9,"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":9,"samples":[{"code_sha256_prefix":"a13a08686e59be7a","entry":"aggregate_metrics","repo":"networkslab/dynn","repo_kind":"official","path":"src/collect_metric_iter.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/collect_metric_iter.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a13a08686e59be7a"}},{"code_sha256_prefix":"88bce24e4c6cd4d0","entry":"calibration_curve","repo":"networkslab/dynn","repo_kind":"official","path":"src/calibration.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/calibration.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"88bce24e4c6cd4d0"}},{"code_sha256_prefix":"f12c8c66097b8183","entry":"compute_conf_threshold","repo":"networkslab/dynn","repo_kind":"official","path":"src/conformal_eedn.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/conformal_eedn.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f12c8c66097b8183"}},{"code_sha256_prefix":"a8d1bf58c8def1e0","entry":"get_conf_thresh","repo":"networkslab/dynn","repo_kind":"official","path":"src/conformal_eedn.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/conformal_eedn.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a8d1bf58c8def1e0"}},{"code_sha256_prefix":"f5fbf06b6a89ad80","entry":"get_display","repo":"networkslab/dynn","repo_kind":"official","path":"src/log_helper.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/log_helper.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f5fbf06b6a89ad80"}},{"code_sha256_prefix":"5372768909366c79","entry":"get_ml_flow_dict","repo":"networkslab/dynn","repo_kind":"official","path":"src/threshold_helper.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/threshold_helper.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5372768909366c79"}},{"code_sha256_prefix":"20b10511bb2127c3","entry":"get_pred_sets","repo":"networkslab/dynn","repo_kind":"official","path":"src/conformal_eedn.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/conformal_eedn.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"20b10511bb2127c3"}},{"code_sha256_prefix":"ad2afb8d830866da","entry":"key_to_name_display","repo":"networkslab/dynn","repo_kind":"official","path":"src/log_helper.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/log_helper.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ad2afb8d830866da"}},{"code_sha256_prefix":"4d7fdafff6eb3a88","entry":"we_want_to_see_it","repo":"networkslab/dynn","repo_kind":"official","path":"src/log_helper.py","file_url":"https://github.com/networkslab/dynn/blob/HEAD/src/log_helper.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4d7fdafff6eb3a88"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}