{"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/a-statistical-framework-for-low-bitwidth","title":"A Statistical Framework for Low-bitwidth Training of Deep Neural Networks","arxiv_id":"2010.14298","date":"2020-10-27","proceeding":"NeurIPS 2020 12","authors":["Jianfei Chen","Yu Gai","Zhewei Yao","Michael W. Mahoney","Joseph E. Gonzalez"],"abstract":"Fully quantized training (FQT), which uses low-bitwidth hardware by quantizing the activations, weights, and gradients of a neural network model, is a promising approach to accelerate the training of deep neural networks. One major challenge with FQT is the lack of theoretical understanding, in particular of how gradient quantization impacts convergence properties. In this paper, we address this problem by presenting a statistical framework for analyzing FQT algorithms. We view the quantized gradient of FQT as a stochastic estimator of its full precision counterpart, a procedure known as quantization-aware training (QAT). We show that the FQT gradient is an unbiased estimator of the QAT gradient, and we discuss the impact of gradient quantization on its variance. Inspired by these theoretical results, we develop two novel gradient quantizers, and we show that these have smaller variance than the existing per-tensor quantizer. For training ResNet-50 on ImageNet, our 5-bit block Householder quantizer achieves only 0.5% validation accuracy loss relative to QAT, comparable to the existing INT8 baseline.","url_abs":"https://arxiv.org/abs/2010.14298v1","url_pdf":"https://arxiv.org/pdf/2010.14298v1.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":"a-statistical-framework-for-low-bitwidth","repo_url":"https://github.com/cjf00000/StatQuant","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"a-statistical-framework-for-low-bitwidth","repo_url":"https://github.com/gaochang-bjtu/1-bit-fqt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"linguistic-acceptability","task_name":"Linguistic Acceptability"},{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"quantization","task_name":"Quantization"},{"task_slug":"semantic-textual-similarity","task_name":"Semantic Textual Similarity"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/linguistic-acceptability-on-cola","task":"Linguistic Acceptability","dataset":"CoLA","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":21,"of":43,"metrics":{"Accuracy":"67.5"},"uses_additional_data":false},{"leaderboard":"/sota/natural-language-inference-on-multinli","task":"Natural Language Inference","dataset":"MultiNLI","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":13,"of":67,"metrics":{"Matched":"89.9"},"uses_additional_data":false},{"leaderboard":"/sota/natural-language-inference-on-qnli","task":"Natural Language Inference","dataset":"QNLI","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":15,"of":43,"metrics":{"Accuracy":"94.5"},"uses_additional_data":false},{"leaderboard":"/sota/natural-language-inference-on-rte","task":"Natural Language Inference","dataset":"RTE","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":23,"of":90,"metrics":{"Accuracy":"86.8"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-textual-similarity-on-mrpc","task":"Semantic Textual Similarity","dataset":"MRPC","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":13,"of":45,"metrics":{"Accuracy":"90.4"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-textual-similarity-on-sts-benchmark","task":"Semantic Textual Similarity","dataset":"STS Benchmark","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":10,"of":66,"metrics":{"Pearson Correlation":"0.919"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-sst-2-binary","task":"Sentiment Analysis","dataset":"SST-2 Binary classification","model":"PSQ (Chen et al., 2020)","rank_in_archive_order":20,"of":87,"metrics":{"Accuracy":"96.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2010.14298","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2010.14298"}},"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/cjf00000/StatQuant","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gaochang-bjtu/1-bit-fqt","reach":null}],"summary":{"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"official":{"samples":3,"ran":1,"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":1,"samples":[{"code_sha256_prefix":"03380fb7e632a2af","entry":"get_optimizer","repo":"cjf00000/StatQuant","repo_kind":"official","path":"image_classification/training.py","file_url":"https://github.com/cjf00000/StatQuant/blob/HEAD/image_classification/training.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"03380fb7e632a2af"}},{"code_sha256_prefix":"2e099f52e7267bec","entry":"lr_policy","repo":"cjf00000/StatQuant","repo_kind":"official","path":"image_classification/training.py","file_url":"https://github.com/cjf00000/StatQuant/blob/HEAD/image_classification/training.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":"2e099f52e7267bec"}},{"code_sha256_prefix":"15d95c6c7625379f","entry":"lr_step_policy","repo":"cjf00000/StatQuant","repo_kind":"official","path":"image_classification/training.py","file_url":"https://github.com/cjf00000/StatQuant/blob/HEAD/image_classification/training.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":"15d95c6c7625379f"}},{"code_sha256_prefix":"cda3783cdf085eef","entry":"validate","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"cda3783cdf085eef"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}