{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/neural-offset-min-sum-decoding","title":"Neural Offset Min-Sum Decoding","arxiv_id":"1701.05931","date":"2017-01-20","proceeding":null,"authors":["Loren Lugosch","Warren J. Gross"],"abstract":"Recently, it was shown that if multiplicative weights are assigned to the\nedges of a Tanner graph used in belief propagation decoding, it is possible to\nuse deep learning techniques to find values for the weights which improve the\nerror-correction performance of the decoder. Unfortunately, this approach\nrequires many multiplications, which are generally expensive operations. In\nthis paper, we suggest a more hardware-friendly approach in which offset\nmin-sum decoding is augmented with learnable offset parameters. Our method uses\nno multiplications and has a parameter count less than half that of the\nmultiplicative algorithm. This both speeds up training and provides a feasible\npath to hardware architectures. After describing our method, we compare the\nperformance of the two neural decoding algorithms and show that our method\nachieves error-correction performance within 0.1 dB of the multiplicative\napproach and as much as 1 dB better than traditional belief propagation for the\ncodes under consideration.","url_abs":"http://arxiv.org/abs/1701.05931v3","url_pdf":"http://arxiv.org/pdf/1701.05931v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"neural-offset-min-sum-decoding","repo_url":"https://github.com/lorenlugosch/neural-min-sum-decoding","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1701.05931","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}