{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:ED4F53GCRLOJRJHSWFUN2OGZG2","short_pith_number":"pith:ED4F53GC","canonical_record":{"source":{"id":"1706.02379","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-07T21:01:15Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"7711ee2d53154a775ca3865648ba900ba99cbcb34830659dcd6da8276996eabf","abstract_canon_sha256":"1d2599d522c37e2bd42ae1fd66338cc9a91fc2fe092587c08458b750564d37c4"},"schema_version":"1.0"},"canonical_sha256":"20f85eecc28adc98a4f2b168dd38d936a9d5aee7351cacb97ca1fdce614fda8f","source":{"kind":"arxiv","id":"1706.02379","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1706.02379","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"arxiv_version","alias_value":"1706.02379v3","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.02379","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"pith_short_12","alias_value":"ED4F53GCRLOJ","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_16","alias_value":"ED4F53GCRLOJRJHS","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_8","alias_value":"ED4F53GC","created_at":"2026-05-18T12:31:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:ED4F53GCRLOJRJHSWFUN2OGZG2","target":"record","payload":{"canonical_record":{"source":{"id":"1706.02379","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-07T21:01:15Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"7711ee2d53154a775ca3865648ba900ba99cbcb34830659dcd6da8276996eabf","abstract_canon_sha256":"1d2599d522c37e2bd42ae1fd66338cc9a91fc2fe092587c08458b750564d37c4"},"schema_version":"1.0"},"canonical_sha256":"20f85eecc28adc98a4f2b168dd38d936a9d5aee7351cacb97ca1fdce614fda8f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:30:44.453125Z","signature_b64":"BzXm3ZrF6elJbtr57zgTlynSrK9bYk8NGTIsfT9fjUbmXh05q6DlRoYyzbEWk0O/vbO+3VFtvAxOAa7R/eR2Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20f85eecc28adc98a4f2b168dd38d936a9d5aee7351cacb97ca1fdce614fda8f","last_reissued_at":"2026-05-18T00:30:44.452550Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:30:44.452550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1706.02379","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:30:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oC6mi677kPFtOvthRmw0IDcSYEDXh/4vcCLI1hNZpT8UbFrsG4uPcssso13+GHKcU1HgSouG0oVg0BwHgIaiDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T23:05:23.718240Z"},"content_sha256":"e10e42ba283256de8a1db32a35b72b9ce4328b0862d88efcbc85b4517aa31c65","schema_version":"1.0","event_id":"sha256:e10e42ba283256de8a1db32a35b72b9ce4328b0862d88efcbc85b4517aa31c65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:ED4F53GCRLOJRJHSWFUN2OGZG2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training Quantized Nets: A Deeper Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Christoph Studer, Hanan Samet, Hao Li, Soham De, Tom Goldstein, Zheng Xu","submitted_at":"2017-06-07T21:01:15Z","abstract_excerpt":"Currently, deep neural networks are deployed on low-power portable devices by first training a full-precision model using powerful hardware, and then deriving a corresponding low-precision model for efficient inference on such systems. However, training models directly with coarsely quantized weights is a key step towards learning on embedded platforms that have limited computing resources, memory capacity, and power consumption. Numerous recent publications have studied methods for training quantized networks, but these studies have mostly been empirical. In this work, we investigate training"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.02379","kind":"arxiv","version":3},"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":"2026-05-18T00:30:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y3fd5SYHr5Tcz5XOKE1116xVkrOhVUVPYcJ4DyUF2nQdvt19029/2/VXDBXgVOoAviVyHYhWvhkCT83HeOA3BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T23:05:23.718627Z"},"content_sha256":"270c714e159f782a190689d99ebfda00c93de9a0e22c0bca51bda186a13c8e5f","schema_version":"1.0","event_id":"sha256:270c714e159f782a190689d99ebfda00c93de9a0e22c0bca51bda186a13c8e5f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ED4F53GCRLOJRJHSWFUN2OGZG2/bundle.json","state_url":"https://pith.science/pith/ED4F53GCRLOJRJHSWFUN2OGZG2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ED4F53GCRLOJRJHSWFUN2OGZG2/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":"2026-06-02T23:05:23Z","links":{"resolver":"https://pith.science/pith/ED4F53GCRLOJRJHSWFUN2OGZG2","bundle":"https://pith.science/pith/ED4F53GCRLOJRJHSWFUN2OGZG2/bundle.json","state":"https://pith.science/pith/ED4F53GCRLOJRJHSWFUN2OGZG2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ED4F53GCRLOJRJHSWFUN2OGZG2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:ED4F53GCRLOJRJHSWFUN2OGZG2","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":"1d2599d522c37e2bd42ae1fd66338cc9a91fc2fe092587c08458b750564d37c4","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-07T21:01:15Z","title_canon_sha256":"7711ee2d53154a775ca3865648ba900ba99cbcb34830659dcd6da8276996eabf"},"schema_version":"1.0","source":{"id":"1706.02379","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1706.02379","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"arxiv_version","alias_value":"1706.02379v3","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.02379","created_at":"2026-05-18T00:30:44Z"},{"alias_kind":"pith_short_12","alias_value":"ED4F53GCRLOJ","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_16","alias_value":"ED4F53GCRLOJRJHS","created_at":"2026-05-18T12:31:12Z"},{"alias_kind":"pith_short_8","alias_value":"ED4F53GC","created_at":"2026-05-18T12:31:12Z"}],"graph_snapshots":[{"event_id":"sha256:270c714e159f782a190689d99ebfda00c93de9a0e22c0bca51bda186a13c8e5f","target":"graph","created_at":"2026-05-18T00:30:44Z","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":"Currently, deep neural networks are deployed on low-power portable devices by first training a full-precision model using powerful hardware, and then deriving a corresponding low-precision model for efficient inference on such systems. However, training models directly with coarsely quantized weights is a key step towards learning on embedded platforms that have limited computing resources, memory capacity, and power consumption. Numerous recent publications have studied methods for training quantized networks, but these studies have mostly been empirical. In this work, we investigate training","authors_text":"Christoph Studer, Hanan Samet, Hao Li, Soham De, Tom Goldstein, Zheng Xu","cross_cats":["cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-07T21:01:15Z","title":"Training Quantized Nets: A Deeper Understanding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.02379","kind":"arxiv","version":3},"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:e10e42ba283256de8a1db32a35b72b9ce4328b0862d88efcbc85b4517aa31c65","target":"record","created_at":"2026-05-18T00:30:44Z","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":"1d2599d522c37e2bd42ae1fd66338cc9a91fc2fe092587c08458b750564d37c4","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-07T21:01:15Z","title_canon_sha256":"7711ee2d53154a775ca3865648ba900ba99cbcb34830659dcd6da8276996eabf"},"schema_version":"1.0","source":{"id":"1706.02379","kind":"arxiv","version":3}},"canonical_sha256":"20f85eecc28adc98a4f2b168dd38d936a9d5aee7351cacb97ca1fdce614fda8f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"20f85eecc28adc98a4f2b168dd38d936a9d5aee7351cacb97ca1fdce614fda8f","first_computed_at":"2026-05-18T00:30:44.452550Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:30:44.452550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BzXm3ZrF6elJbtr57zgTlynSrK9bYk8NGTIsfT9fjUbmXh05q6DlRoYyzbEWk0O/vbO+3VFtvAxOAa7R/eR2Bg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:30:44.453125Z","signed_message":"canonical_sha256_bytes"},"source_id":"1706.02379","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e10e42ba283256de8a1db32a35b72b9ce4328b0862d88efcbc85b4517aa31c65","sha256:270c714e159f782a190689d99ebfda00c93de9a0e22c0bca51bda186a13c8e5f"],"state_sha256":"5da097edf1dcd8e8be2331d3e64522981cfaf9afad2199f23bba92070627ddae"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fuILKTyuPgqfu/3gRNrvaJeFMui44SqTGMmVwNJ88fXKAX4MK+Hd+nqTL+jqpB2gAu5EwuUgZHpyiVOZakN/CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-02T23:05:23.720571Z","bundle_sha256":"6254d2c756b0811d6273e270cacafce004bcd4c886909b52d294e1cbd7fb183a"}}