{"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/optimistic-verifiable-training-by-controlling","title":"Optimistic Verifiable Training by Controlling Hardware Nondeterminism","arxiv_id":"2403.09603","date":"2024-03-14","proceeding":null,"authors":["Megha Srivastava","Simran Arora","Dan Boneh"],"abstract":"The increasing compute demands of AI systems have led to the emergence of services that train models on behalf of clients lacking necessary resources. However, ensuring correctness of training and guarding against potential training-time attacks, such as data poisoning and backdoors, poses challenges. Existing works on verifiable training largely fall into two classes: proof-based systems, which are difficult to scale, and ``optimistic'' methods that consider a third-party auditor who can replicate the training process and contest the trainer. A key challenge with the latter is that nondeterminism between GPU types during training prevents exact replication of the training process, resulting in schemes that are non-robust. We propose a method that combines training in a higher precision than the target, rounding after intermediate computations, and sharing rounding decisions based on an adaptive thresholding procedure, to successfully control for nondeterminism. Across three different NVIDIA GPUs (A40, Titan XP, RTX 2080 Ti), we achieve exact training replication at FP32 precision for both full-training and fine-tuning of ResNet-50 (23M) and GPT-2 (117M) models. Our verifiable training scheme significantly decreases the storage and time costs compared to proof-based systems, and is publicly released at https://github.com/meghabyte/verifiable-training.","url_abs":"https://arxiv.org/abs/2403.09603v3","url_pdf":"https://arxiv.org/pdf/2403.09603v3.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":"optimistic-verifiable-training-by-controlling","repo_url":"https://github.com/meghabyte/verifiable-training","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"data-poisoning","task_name":"Data Poisoning"},{"task_slug":null,"task_name":"GPU"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-2","method_name":"GPT-2"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.09603","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.09603"}},"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":"deterministic:regex_extraction","url":"https://github.com/meghabyte/verifiable-training","reach":null}],"summary":{"ran":3,"ran_draft_wrong":4,"ran_honours":1,"ran_fixture":1},"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":"31f8cba36a7981f2","entry":"NextAfter","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"round_utils.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/round_utils.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":"31f8cba36a7981f2"}},{"code_sha256_prefix":"3c759b148dad8041","entry":"RoundStandard","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"round_utils.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/round_utils.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":"3c759b148dad8041"}},{"code_sha256_prefix":"f23a02d4e35c0919","entry":"Rounder","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"round_utils.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/round_utils.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":"f23a02d4e35c0919"}},{"code_sha256_prefix":"45c4eeda8c45592f","entry":"apply_nextafter_n","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"round_utils.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/round_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"45c4eeda8c45592f"}},{"code_sha256_prefix":"6163d8e1e3a9ddfb","entry":"create_str","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"merkle.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/merkle.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6163d8e1e3a9ddfb"}},{"code_sha256_prefix":"7ed27cc89e9a22d6","entry":"get_list_leaves","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"merkle.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/merkle.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7ed27cc89e9a22d6"}},{"code_sha256_prefix":"ed82b73e5eb5378c","entry":"round_32","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"round_utils.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/round_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ed82b73e5eb5378c"}},{"code_sha256_prefix":"e56cec769a8fc897","entry":"round_any_final","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"round_utils.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/round_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e56cec769a8fc897"}},{"code_sha256_prefix":"98749e1de6e2db3e","entry":"tensor_to_numpy","repo":"meghabyte/verifiable-training","repo_kind":"official","path":"merkle.py","file_url":"https://github.com/meghabyte/verifiable-training/blob/HEAD/merkle.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"98749e1de6e2db3e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}