{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:KHJD2VHHSNB7LRQ7NHC6YCO5EM","short_pith_number":"pith:KHJD2VHH","canonical_record":{"source":{"id":"1810.10836","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.ET","submitted_at":"2018-10-25T11:49:24Z","cross_cats_sorted":["physics.app-ph"],"title_canon_sha256":"06debcc7760f2b01dc6aa4daa61811f5dbc7386ccc7954af1479f69cfd8404ca","abstract_canon_sha256":"30fc9b0fde957c54c13e1aad61094dee7f5cb04976b7996687d5d5c4126f835e"},"schema_version":"1.0"},"canonical_sha256":"51d23d54e79343f5c61f69c5ec09dd2309e91dc5bbabb2ce336b02229fddefd5","source":{"kind":"arxiv","id":"1810.10836","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.10836","created_at":"2026-05-18T00:02:19Z"},{"alias_kind":"arxiv_version","alias_value":"1810.10836v1","created_at":"2026-05-18T00:02:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.10836","created_at":"2026-05-18T00:02:19Z"},{"alias_kind":"pith_short_12","alias_value":"KHJD2VHHSNB7","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KHJD2VHHSNB7LRQ7","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KHJD2VHH","created_at":"2026-05-18T12:32:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:KHJD2VHHSNB7LRQ7NHC6YCO5EM","target":"record","payload":{"canonical_record":{"source":{"id":"1810.10836","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.ET","submitted_at":"2018-10-25T11:49:24Z","cross_cats_sorted":["physics.app-ph"],"title_canon_sha256":"06debcc7760f2b01dc6aa4daa61811f5dbc7386ccc7954af1479f69cfd8404ca","abstract_canon_sha256":"30fc9b0fde957c54c13e1aad61094dee7f5cb04976b7996687d5d5c4126f835e"},"schema_version":"1.0"},"canonical_sha256":"51d23d54e79343f5c61f69c5ec09dd2309e91dc5bbabb2ce336b02229fddefd5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:02:19.045215Z","signature_b64":"Rns4SI/40nslocWl1zyGiWwWS4aCyWlntYjNvmFkilgWhp1WvYKthwdVi7kd1iNKMcI+XQuds09DajkETKE7AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51d23d54e79343f5c61f69c5ec09dd2309e91dc5bbabb2ce336b02229fddefd5","last_reissued_at":"2026-05-18T00:02:19.044864Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:02:19.044864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1810.10836","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:02:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hbSjRaevBy18B/zKtrhy16xTmpkWKzQTrO9wyBpvGq6DSlFKri2H81E4acA37B6qEbZ6LRh47vlAU0cGbJ3oDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T02:57:05.840951Z"},"content_sha256":"0ef73b4736ce397030d9799e7a24b8a7c9aae7f537665c704a3d1b214c0ce349","schema_version":"1.0","event_id":"sha256:0ef73b4736ce397030d9799e7a24b8a7c9aae7f537665c704a3d1b214c0ce349"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:KHJD2VHHSNB7LRQ7NHC6YCO5EM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Use of Magnetoresistive Random-Access Memory as Approximate Memory for Training Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["physics.app-ph"],"primary_cat":"cs.ET","authors_text":"Adrien F. Vincent, Damien Querlioz, Nicolas Locatelli","submitted_at":"2018-10-25T11:49:24Z","abstract_excerpt":"Hardware neural networks that implement synaptic weights with embedded non-volatile memory, such as spin torque memory (ST-MRAM), are a major lead for low energy artificial intelligence. In this work, we propose an approximate storage approach for their memory. We show that this strategy grants effective control of the bit error rate by modulating the programming pulse amplitude or duration. Accounting for the devices variability issue, we evaluate energy savings, and show how they translate when training a hardware neural network. On an image recognition example, 74% of programming energy can"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.10836","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:02:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qalaMztARsyFUf8hut8TuX7n24fRYP8l8xk0joqFS3k7UwNxUXFrZrnGc1e0R76VgdUMW6NuySwTleeF5g9sBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T02:57:05.841654Z"},"content_sha256":"4df33233a22dda5b88904b9d70b7761d62dca5b0f72ed37f60c4ce20457ce45e","schema_version":"1.0","event_id":"sha256:4df33233a22dda5b88904b9d70b7761d62dca5b0f72ed37f60c4ce20457ce45e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM/bundle.json","state_url":"https://pith.science/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-22T02:57:05Z","links":{"resolver":"https://pith.science/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM","bundle":"https://pith.science/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM/bundle.json","state":"https://pith.science/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KHJD2VHHSNB7LRQ7NHC6YCO5EM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:KHJD2VHHSNB7LRQ7NHC6YCO5EM","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"30fc9b0fde957c54c13e1aad61094dee7f5cb04976b7996687d5d5c4126f835e","cross_cats_sorted":["physics.app-ph"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.ET","submitted_at":"2018-10-25T11:49:24Z","title_canon_sha256":"06debcc7760f2b01dc6aa4daa61811f5dbc7386ccc7954af1479f69cfd8404ca"},"schema_version":"1.0","source":{"id":"1810.10836","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1810.10836","created_at":"2026-05-18T00:02:19Z"},{"alias_kind":"arxiv_version","alias_value":"1810.10836v1","created_at":"2026-05-18T00:02:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.10836","created_at":"2026-05-18T00:02:19Z"},{"alias_kind":"pith_short_12","alias_value":"KHJD2VHHSNB7","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KHJD2VHHSNB7LRQ7","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KHJD2VHH","created_at":"2026-05-18T12:32:33Z"}],"graph_snapshots":[{"event_id":"sha256:4df33233a22dda5b88904b9d70b7761d62dca5b0f72ed37f60c4ce20457ce45e","target":"graph","created_at":"2026-05-18T00:02:19Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Hardware neural networks that implement synaptic weights with embedded non-volatile memory, such as spin torque memory (ST-MRAM), are a major lead for low energy artificial intelligence. In this work, we propose an approximate storage approach for their memory. We show that this strategy grants effective control of the bit error rate by modulating the programming pulse amplitude or duration. Accounting for the devices variability issue, we evaluate energy savings, and show how they translate when training a hardware neural network. On an image recognition example, 74% of programming energy can","authors_text":"Adrien F. Vincent, Damien Querlioz, Nicolas Locatelli","cross_cats":["physics.app-ph"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.ET","submitted_at":"2018-10-25T11:49:24Z","title":"Use of Magnetoresistive Random-Access Memory as Approximate Memory for Training Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.10836","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0ef73b4736ce397030d9799e7a24b8a7c9aae7f537665c704a3d1b214c0ce349","target":"record","created_at":"2026-05-18T00:02:19Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"30fc9b0fde957c54c13e1aad61094dee7f5cb04976b7996687d5d5c4126f835e","cross_cats_sorted":["physics.app-ph"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.ET","submitted_at":"2018-10-25T11:49:24Z","title_canon_sha256":"06debcc7760f2b01dc6aa4daa61811f5dbc7386ccc7954af1479f69cfd8404ca"},"schema_version":"1.0","source":{"id":"1810.10836","kind":"arxiv","version":1}},"canonical_sha256":"51d23d54e79343f5c61f69c5ec09dd2309e91dc5bbabb2ce336b02229fddefd5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"51d23d54e79343f5c61f69c5ec09dd2309e91dc5bbabb2ce336b02229fddefd5","first_computed_at":"2026-05-18T00:02:19.044864Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:02:19.044864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Rns4SI/40nslocWl1zyGiWwWS4aCyWlntYjNvmFkilgWhp1WvYKthwdVi7kd1iNKMcI+XQuds09DajkETKE7AA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:02:19.045215Z","signed_message":"canonical_sha256_bytes"},"source_id":"1810.10836","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0ef73b4736ce397030d9799e7a24b8a7c9aae7f537665c704a3d1b214c0ce349","sha256:4df33233a22dda5b88904b9d70b7761d62dca5b0f72ed37f60c4ce20457ce45e"],"state_sha256":"254267d2f26345d91c073e183a975197544a3b625af58777b6ba31df7f1ffbca"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UTqMzU4uZtc6LbcajwtuPM8hzgKlL5haXr32trxi2M9Nlc0iHupMdFG/n1XyOA2xsEig4ZwCuuAe5LfyhvahAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-22T02:57:05.845072Z","bundle_sha256":"3f2dc93515b6b22112f9601a2cbbc2ac96138b570ed3176ee0d34a6831a646e2"}}