{"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/monarch-expressive-structured-matrices-for","title":"Monarch: Expressive Structured Matrices for Efficient and Accurate Training","arxiv_id":"2204.00595","date":"2022-04-01","proceeding":null,"authors":["Tri Dao","Beidi Chen","Nimit Sohoni","Arjun Desai","Michael Poli","Jessica Grogan","Alexander Liu","Aniruddh Rao","Atri Rudra","Christopher Ré"],"abstract":"Large neural networks excel in many domains, but they are expensive to train and fine-tune. A popular approach to reduce their compute or memory requirements is to replace dense weight matrices with structured ones (e.g., sparse, low-rank, Fourier transform). These methods have not seen widespread adoption (1) in end-to-end training due to unfavorable efficiency--quality tradeoffs, and (2) in dense-to-sparse fine-tuning due to lack of tractable algorithms to approximate a given dense weight matrix. To address these issues, we propose a class of matrices (Monarch) that is hardware-efficient (they are parameterized as products of two block-diagonal matrices for better hardware utilization) and expressive (they can represent many commonly used transforms). Surprisingly, the problem of approximating a dense weight matrix with a Monarch matrix, though nonconvex, has an analytical optimal solution. These properties of Monarch matrices unlock new ways to train and fine-tune sparse and dense models. We empirically validate that Monarch can achieve favorable accuracy-efficiency tradeoffs in several end-to-end sparse training applications: speeding up ViT and GPT-2 training on ImageNet classification and Wikitext-103 language modeling by 2x with comparable model quality, and reducing the error on PDE solving and MRI reconstruction tasks by 40%. In sparse-to-dense training, with a simple technique called \"reverse sparsification,\" Monarch matrices serve as a useful intermediate representation to speed up GPT-2 pretraining on OpenWebText by 2x without quality drop. The same technique brings 23% faster BERT pretraining than even the very optimized implementation from Nvidia that set the MLPerf 1.1 record. In dense-to-sparse fine-tuning, as a proof-of-concept, our Monarch approximation algorithm speeds up BERT fine-tuning on GLUE by 1.7x with comparable accuracy.","url_abs":"https://arxiv.org/abs/2204.00595v1","url_pdf":"https://arxiv.org/pdf/2204.00595v1.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":"monarch-expressive-structured-matrices-for","repo_url":"https://github.com/hazyresearch/monarch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"monarch-expressive-structured-matrices-for","repo_url":"https://github.com/pascalcarrivain/ksmm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"mri-reconstruction","task_name":"MRI Reconstruction"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"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":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2204.00595","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2204.00595"}},"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/pascalcarrivain/ksmm","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hazyresearch/monarch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_violates":1,"ran_fixture":1,"unverified":13},"by_repo_kind":{"listed":{"samples":15,"ran":2,"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":"6ba8cee9f5daea41","entry":"pair","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/model/vit.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/model/vit.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6ba8cee9f5daea41"}},{"code_sha256_prefix":"17f6d223949a7e58","entry":"posemb_sincos_2d","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/model/vit.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/model/vit.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"17f6d223949a7e58"}},{"code_sha256_prefix":"28185b31217241b1","entry":"add_lines_to_instantiate_kernel_with_hp","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/layer/kronecker_sparse/kernel.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/layer/kronecker_sparse/kernel.py","link_basis":"harvester_set","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":"28185b31217241b1"}},{"code_sha256_prefix":"500510d20d2cda99","entry":"forward_bmm_bs_first","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/layer/bmm/forward_bmm.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/layer/bmm/forward_bmm.py","link_basis":"harvester_set","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":"500510d20d2cda99"}},{"code_sha256_prefix":"59f540493105da3a","entry":"forward_bmm_bs_last","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/layer/bmm/forward_bmm.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/layer/bmm/forward_bmm.py","link_basis":"harvester_set","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":"59f540493105da3a"}},{"code_sha256_prefix":"64f5feef032ab239","entry":"forward_pass","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/benchmark/utils.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/benchmark/utils.py","link_basis":"harvester_set","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":"64f5feef032ab239"}},{"code_sha256_prefix":"2248676f55f31017","entry":"from_bmm_format_to_dense","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/layer/kronecker_sparse/interface.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/layer/kronecker_sparse/interface.py","link_basis":"harvester_set","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":"2248676f55f31017"}},{"code_sha256_prefix":"5889d0ad88e0d5f5","entry":"from_kernel_format_to_dense","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/layer/kronecker_sparse/interface.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/layer/kronecker_sparse/interface.py","link_basis":"harvester_set","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":"5889d0ad88e0d5f5"}},{"code_sha256_prefix":"d43163d9a5d6f145","entry":"generate_factor","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/layer/kronecker_sparse/interface.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/layer/kronecker_sparse/interface.py","link_basis":"harvester_set","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":"d43163d9a5d6f145"}},{"code_sha256_prefix":"49146715588712c1","entry":"get_input_size","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_cu/long_run.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_cu/long_run.py","link_basis":"harvester_set","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":"49146715588712c1"}},{"code_sha256_prefix":"90795aa963bed427","entry":"is_gpu_available","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_cu/long_run.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_cu/long_run.py","link_basis":"harvester_set","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":"90795aa963bed427"}},{"code_sha256_prefix":"f201b7609964135d","entry":"is_valid_configuration_sanity_check","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_cu/long_run.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_cu/long_run.py","link_basis":"harvester_set","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":"f201b7609964135d"}},{"code_sha256_prefix":"8358b5b9e35e1a5f","entry":"pattern_is_dense","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/benchmark/utils.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/benchmark/utils.py","link_basis":"harvester_set","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":"8358b5b9e35e1a5f"}},{"code_sha256_prefix":"a076f484cbf3ed79","entry":"posemb_sincos_2d_bsl","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/model/vit.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/model/vit.py","link_basis":"harvester_set","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":"a076f484cbf3ed79"}},{"code_sha256_prefix":"ab4fa31231627e50","entry":"set_device_and_get_device_name","repo":"pascalcarrivain/ksmm","repo_kind":"listed","path":"src/ksmm_py/benchmark/utils.py","file_url":"https://github.com/pascalcarrivain/ksmm/blob/HEAD/src/ksmm_py/benchmark/utils.py","link_basis":"harvester_set","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":"ab4fa31231627e50"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}