{"schema_version":"1.0","canonical_url":"https://patentable.app/patents/US-11468370","patent":{"patent_number":"US-11468370","title":"Communication compression method based on model weight distribution in federated learning","assignee":null,"inventors":[],"filing_date":"2022-03-07T00:00:00.000Z","publication_date":"2022-10-11T00:00:00.000Z","cpc_codes":["G06N","G06F","G06N","G06N","G06N","G06N"],"num_claims":4,"abstract":"A communication compression method based on model weight distribution in federated learning, and belongs to the technical field of wireless communication. Based on the existing federated average idea in federated learning, counts the distribution of model weight information to be transmitted between nodes during each communication, then performs scalar quantization and compression through Lloyd-Max quantizer according to their distribution characteristics, then encodes with Huffman coding method, and finally sends the codes to the target node, thereby the minimum mean square quantization error is realized and the number of bits required for communication is reduced."},"analysis":{"summary":null,"layman_explanation":null,"technical_analysis":null,"business_analysis":null,"faqs":null,"topics":[],"tech_cluster":null},"seo":{"title":"Communication compression method based on model weight distribution in federated learning","description":"A communication compression method based on model weight distribution in federated learning, and belongs to the technical field of wireless communication. Based on the existing federated average idea ","keywords":[]},"attribution":{"source":"Patentable","source_url":"https://patentable.app","canonical_url":"https://patentable.app/patents/US-11468370","license":"CC-BY-4.0-like","license_terms":"AI-generated analysis on this page (summary, layman_explanation, technical_analysis, business_analysis, faqs) may be reused with attribution and a visible link back to the canonical URL above. Patent abstracts, claims, and bibliographic data are USPTO public domain.","required_link":"https://patentable.app/patents/US-11468370","citation_suggestion":"Patentable. \"Communication compression method based on model weight distribution in federated learning\" (US-11468370). https://patentable.app/patents/US-11468370","copyright_holder":"Nomic Interactive Technology LLC"},"links":{"html":"https://patentable.app/patents/US-11468370","json":"https://patentable.app/api/llm-context/US-11468370","site":"https://patentable.app","llms_txt":"https://patentable.app/llms.txt"},"generated_at":"2026-05-30T17:16:52.253Z"}