{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:RNXIQ76L7PWOO7SKRICTSEDYRT","short_pith_number":"pith:RNXIQ76L","schema_version":"1.0","canonical_sha256":"8b6e887fcbfbece77e4a8a053910788ce2f3ace3f2cc695ad69f77f06798fb6f","source":{"kind":"arxiv","id":"2103.04003","version":1},"attestation_state":"computed","paper":{"title":"Memory-efficient Learning for High-Dimensional MRI Reconstruction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"eess.IV","authors_text":"Christopher M. Sandino, Jonathan I. Tamir, Kevin Zhang, Ke Wang, Michael Kellman, Michael Lustig, Shreyas S. Vasanawala, Stella X. Yu","submitted_at":"2021-03-06T01:36:25Z","abstract_excerpt":"Deep learning (DL) based unrolled reconstructions have shown state-of-the-art performance for under-sampled magnetic resonance imaging (MRI). Similar to compressed sensing, DL can leverage high-dimensional data (e.g. 3D, 2D+time, 3D+time) to further improve performance. However, network size and depth are currently limited by the GPU memory required for backpropagation. Here we use a memory-efficient learning (MEL) framework which favorably trades off storage with a manageable increase in computation during training. Using MEL with multi-dimensional data, we demonstrate improved image reconstr"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2103.04003","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.IV","submitted_at":"2021-03-06T01:36:25Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"cef245ecf84ce2a6964e901e185f57b05c841c4bc51d7e1f8d23e3a696bae57a","abstract_canon_sha256":"1e2cfcafa14334ce7b1c62712bb9aa4f8ec792d1967afe9c474477096dccc5ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:20:44.900058Z","signature_b64":"ydv6gVDB5i+xI9TOa3hjmqrVHl7P9vEaORsoets4+lY8d31DNwnHYfObrjPEHnCoNkNJTmsu196W8dnS1UyZBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b6e887fcbfbece77e4a8a053910788ce2f3ace3f2cc695ad69f77f06798fb6f","last_reissued_at":"2026-07-05T02:20:44.899704Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:20:44.899704Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memory-efficient Learning for High-Dimensional MRI Reconstruction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"eess.IV","authors_text":"Christopher M. Sandino, Jonathan I. Tamir, Kevin Zhang, Ke Wang, Michael Kellman, Michael Lustig, Shreyas S. Vasanawala, Stella X. Yu","submitted_at":"2021-03-06T01:36:25Z","abstract_excerpt":"Deep learning (DL) based unrolled reconstructions have shown state-of-the-art performance for under-sampled magnetic resonance imaging (MRI). Similar to compressed sensing, DL can leverage high-dimensional data (e.g. 3D, 2D+time, 3D+time) to further improve performance. However, network size and depth are currently limited by the GPU memory required for backpropagation. Here we use a memory-efficient learning (MEL) framework which favorably trades off storage with a manageable increase in computation during training. Using MEL with multi-dimensional data, we demonstrate improved image reconstr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.04003","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2103.04003/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2103.04003","created_at":"2026-07-05T02:20:44.899768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.04003v1","created_at":"2026-07-05T02:20:44.899768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.04003","created_at":"2026-07-05T02:20:44.899768+00:00"},{"alias_kind":"pith_short_12","alias_value":"RNXIQ76L7PWO","created_at":"2026-07-05T02:20:44.899768+00:00"},{"alias_kind":"pith_short_16","alias_value":"RNXIQ76L7PWOO7SK","created_at":"2026-07-05T02:20:44.899768+00:00"},{"alias_kind":"pith_short_8","alias_value":"RNXIQ76L","created_at":"2026-07-05T02:20:44.899768+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT","json":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT.json","graph_json":"https://pith.science/api/pith-number/RNXIQ76L7PWOO7SKRICTSEDYRT/graph.json","events_json":"https://pith.science/api/pith-number/RNXIQ76L7PWOO7SKRICTSEDYRT/events.json","paper":"https://pith.science/paper/RNXIQ76L"},"agent_actions":{"view_html":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT","download_json":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT.json","view_paper":"https://pith.science/paper/RNXIQ76L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.04003&json=true","fetch_graph":"https://pith.science/api/pith-number/RNXIQ76L7PWOO7SKRICTSEDYRT/graph.json","fetch_events":"https://pith.science/api/pith-number/RNXIQ76L7PWOO7SKRICTSEDYRT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT/action/storage_attestation","attest_author":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT/action/author_attestation","sign_citation":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT/action/citation_signature","submit_replication":"https://pith.science/pith/RNXIQ76L7PWOO7SKRICTSEDYRT/action/replication_record"}},"created_at":"2026-07-05T02:20:44.899768+00:00","updated_at":"2026-07-05T02:20:44.899768+00:00"}