{"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/randomness-in-neural-network-training","title":"Randomness In Neural Network Training: Characterizing The Impact of Tooling","arxiv_id":"2106.11872","date":"2021-06-22","proceeding":null,"authors":["Donglin Zhuang","Xingyao Zhang","Shuaiwen Leon Song","Sara Hooker"],"abstract":"The quest for determinism in machine learning has disproportionately focused on characterizing the impact of noise introduced by algorithmic design choices. In this work, we address a less well understood and studied question: how does our choice of tooling introduce randomness to deep neural network training. We conduct large scale experiments across different types of hardware, accelerators, state of art networks, and open-source datasets, to characterize how tooling choices contribute to the level of non-determinism in a system, the impact of said non-determinism, and the cost of eliminating different sources of noise. Our findings are surprising, and suggest that the impact of non-determinism in nuanced. While top-line metrics such as top-1 accuracy are not noticeably impacted, model performance on certain parts of the data distribution is far more sensitive to the introduction of randomness. Our results suggest that deterministic tooling is critical for AI safety. However, we also find that the cost of ensuring determinism varies dramatically between neural network architectures and hardware types, e.g., with overhead up to $746\\%$, $241\\%$, and $196\\%$ on a spectrum of widely used GPU accelerator architectures, relative to non-deterministic training. The source code used in this paper is available at https://github.com/usyd-fsalab/NeuralNetworkRandomness.","url_abs":"https://arxiv.org/abs/2106.11872v1","url_pdf":"https://arxiv.org/pdf/2106.11872v1.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":"randomness-in-neural-network-training","repo_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":null,"task_name":"GPU"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.11872","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.11872"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/usyd-fsalab/NeuralNetworkRandomness","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":14},"by_repo_kind":{"official":{"samples":14,"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":"b587e238a34a5ea6","entry":"MediumCNN","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/cnns.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/cnns.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b587e238a34a5ea6"}},{"code_sha256_prefix":"11da740fe6395a4a","entry":"ResidualBlock","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/resnet.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/resnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"11da740fe6395a4a"}},{"code_sha256_prefix":"7ceef7834e740164","entry":"celebA_transform","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/training_script/resnet_celeba.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/training_script/resnet_celeba.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7ceef7834e740164"}},{"code_sha256_prefix":"52b7a2ba9ec2f57c","entry":"cifar_data_preprocessing","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/input_pipeline.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/input_pipeline.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"52b7a2ba9ec2f57c"}},{"code_sha256_prefix":"6ec7fa06967e3f07","entry":"cifar_random_crop_and_flip","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/input_pipeline.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/input_pipeline.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6ec7fa06967e3f07"}},{"code_sha256_prefix":"98b375ca4e1053fc","entry":"conv2d_bn","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/resnet.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/resnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"98b375ca4e1053fc"}},{"code_sha256_prefix":"be3aecbe3c13892c","entry":"conv2d_bn_relu","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/resnet.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/resnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"be3aecbe3c13892c"}},{"code_sha256_prefix":"e120ab8759b2766a","entry":"get_standard_deviation_of_accuracy","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"run_ae.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/run_ae.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e120ab8759b2766a"}},{"code_sha256_prefix":"7e9d8b95a3c7afeb","entry":"get_standard_deviation_of_fnr","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"run_ae.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/run_ae.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7e9d8b95a3c7afeb"}},{"code_sha256_prefix":"b990fe5855fdceff","entry":"get_standard_deviation_of_fpr","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"run_ae.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/run_ae.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b990fe5855fdceff"}},{"code_sha256_prefix":"eafcda6bb13a0e7c","entry":"horizontal_flip","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/common/input_pipeline.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/common/input_pipeline.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"eafcda6bb13a0e7c"}},{"code_sha256_prefix":"9f3fb2653030c316","entry":"lr_schedule","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/training_script/smallcnn.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/training_script/smallcnn.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9f3fb2653030c316"}},{"code_sha256_prefix":"384218efc7e9fdd8","entry":"lr_scheduler","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/training_script/resnet_celeba.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/training_script/resnet_celeba.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"384218efc7e9fdd8"}},{"code_sha256_prefix":"0c8bdadaba87adfd","entry":"preprocess_image","repo":"usyd-fsalab/NeuralNetworkRandomness","repo_kind":"official","path":"src/training_script/smallcnn.py","file_url":"https://github.com/usyd-fsalab/NeuralNetworkRandomness/blob/HEAD/src/training_script/smallcnn.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0c8bdadaba87adfd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}