{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MBB4B7VMIUQHJOBTQAA3W2HR5Z","short_pith_number":"pith:MBB4B7VM","schema_version":"1.0","canonical_sha256":"6043c0feac452074b8338001bb68f1ee5e65dd466e20733753c349d5b0125c06","source":{"kind":"arxiv","id":"2212.13345","version":1},"attestation_state":"computed","paper":{"title":"The Forward-Forward Algorithm: Some Preliminary Investigations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Geoffrey Hinton","submitted_at":"2022-12-27T02:54:46Z","abstract_excerpt":"The aim of this paper is to introduce a new learning procedure for neural networks and to demonstrate that it works well enough on a few small problems to be worth further investigation. The Forward-Forward algorithm replaces the forward and backward passes of backpropagation by two forward passes, one with positive (i.e. real) data and the other with negative data which could be generated by the network itself. Each layer has its own objective function which is simply to have high goodness for positive data and low goodness for negative data. The sum of the squared activities in a layer can b"},"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":"2212.13345","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-12-27T02:54:46Z","cross_cats_sorted":[],"title_canon_sha256":"20312e01d27bb10cce4542ba2dd4aad00bdac35266cd8678915a4cb201c17b00","abstract_canon_sha256":"09ee603b3ec62e67288d84628794bb0c1175a0410aad0e60d8e2db3b1ab93312"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:28:50.658473Z","signature_b64":"XddkZM+uXm6t+GNVyoY/b3OW+3h8WZUpE4MCKvVsFcqZx8bPBAkJY8UqNqmKRnBAU+he8rMVJ/88i7Zoexs2DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6043c0feac452074b8338001bb68f1ee5e65dd466e20733753c349d5b0125c06","last_reissued_at":"2026-07-05T05:28:50.658052Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:28:50.658052Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Forward-Forward Algorithm: Some Preliminary Investigations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Geoffrey Hinton","submitted_at":"2022-12-27T02:54:46Z","abstract_excerpt":"The aim of this paper is to introduce a new learning procedure for neural networks and to demonstrate that it works well enough on a few small problems to be worth further investigation. The Forward-Forward algorithm replaces the forward and backward passes of backpropagation by two forward passes, one with positive (i.e. real) data and the other with negative data which could be generated by the network itself. Each layer has its own objective function which is simply to have high goodness for positive data and low goodness for negative data. The sum of the squared activities in a layer can b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.13345","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/2212.13345/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":"2212.13345","created_at":"2026-07-05T05:28:50.658113+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.13345v1","created_at":"2026-07-05T05:28:50.658113+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.13345","created_at":"2026-07-05T05:28:50.658113+00:00"},{"alias_kind":"pith_short_12","alias_value":"MBB4B7VMIUQH","created_at":"2026-07-05T05:28:50.658113+00:00"},{"alias_kind":"pith_short_16","alias_value":"MBB4B7VMIUQHJOBT","created_at":"2026-07-05T05:28:50.658113+00:00"},{"alias_kind":"pith_short_8","alias_value":"MBB4B7VM","created_at":"2026-07-05T05:28:50.658113+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":37,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08406","citing_title":"Beyond Backpropagation: Monte Carlo Method Can Train Deep Neural Networks","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2606.24722","citing_title":"Decentralised AI Training and Inference with BlockTrain","ref_index":86,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26515","citing_title":"Forget, Anticipate and Adapt: Test Time Training for Long Videos","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23840","citing_title":"Embodied Explainability and Ontological Obstacles: Why We Struggle to Explain the Answers of Large Language Models (LLMs)","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21497","citing_title":"Breaking chains with trees: Deep learning with $\\mathcal{O}(\\log N)$ parallel time complexity","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21126","citing_title":"What Accuracy and Gradient Cosine Miss: Evaluating Feedback Alignment via Scale Stability, Reference Validity, and Depth Utility","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09734","citing_title":"Adaptive directional gradients for parameterised quantum circuits","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09928","citing_title":"Forward-Only Convolutional Neural Networks with Learnable Channel-Class Assignment","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06424","citing_title":"Intrinsic Computational Functionalism: From Observer-Relative Maps to Observer-Independent Structures","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06539","citing_title":"Synthetic Benchmarks Overstate Forward-Forward Scaling: Real-Data Limits of Layer-Local Training","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03927","citing_title":"FFR: Forward-Forward Learning for Regression","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24797","citing_title":"HCL-FF: Hierarchical and Contrastive Learning for Forward-Forward Algorithm","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26515","citing_title":"Forget, Anticipate and Adapt: Test Time Training for Long Videos","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25062","citing_title":"Cultivating Machine Intelligence: The OMEGA Shift from Top-Down Optimization to Autopoietic Cognitive Ecologies","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27946","citing_title":"Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30638","citing_title":"Score Broadcast and Decorrelation: A General Framework for Broadcast-Based Credit Assignment","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06424","citing_title":"Intrinsic Computational Functionalism: From Observer-Relative Maps to Observer-Independent Structures","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.25627","citing_title":"TL++: Accuracy and Privacy Preserving Traversal Learning for Distributed Intelligent Systems","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2304.01844","citing_title":"Grid-SD2E: A General Grid-Feedback in a System for Cognitive Learning","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2402.14598","citing_title":"MemFlow: A Lightweight Forward Memorizing Framework for Quick Domain Adaptive Feature Mapping","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12406","citing_title":"LightTune: Lightweight Forward-Only Online Fine-Tuning with Applications to Link Adaptation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18804","citing_title":"Adaptive Multi-Scale Goodness Aggregation for Forward-Forward Learning","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15622","citing_title":"Position: Zeroth-Order Optimization in Deep Learning Is Underexplored, Not Underpowered","ref_index":112,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18557","citing_title":"Self-supervised local learning rules learn the hidden hierarchical structure of high-dimensional data","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16872","citing_title":"Some[Body] Must Receive That Pain for Agent Accountability","ref_index":76,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z","json":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z.json","graph_json":"https://pith.science/api/pith-number/MBB4B7VMIUQHJOBTQAA3W2HR5Z/graph.json","events_json":"https://pith.science/api/pith-number/MBB4B7VMIUQHJOBTQAA3W2HR5Z/events.json","paper":"https://pith.science/paper/MBB4B7VM"},"agent_actions":{"view_html":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z","download_json":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z.json","view_paper":"https://pith.science/paper/MBB4B7VM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.13345&json=true","fetch_graph":"https://pith.science/api/pith-number/MBB4B7VMIUQHJOBTQAA3W2HR5Z/graph.json","fetch_events":"https://pith.science/api/pith-number/MBB4B7VMIUQHJOBTQAA3W2HR5Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z/action/storage_attestation","attest_author":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z/action/author_attestation","sign_citation":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z/action/citation_signature","submit_replication":"https://pith.science/pith/MBB4B7VMIUQHJOBTQAA3W2HR5Z/action/replication_record"}},"created_at":"2026-07-05T05:28:50.658113+00:00","updated_at":"2026-07-05T05:28:50.658113+00:00"}