{"id":"https://openalex.org/W7128527422","doi":"https://doi.org/10.48550/arxiv.2602.07400","title":"BitLogic: Training Framework for Gradient-Based FPGA-Native Neural Networks","display_name":"BitLogic: Training Framework for Gradient-Based FPGA-Native Neural Networks","publication_year":2026,"publication_date":"2026-02-07","ids":{"openalex":"https://openalex.org/W7128527422","doi":"https://doi.org/10.48550/arxiv.2602.07400"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.07400","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119805118","display_name":"Simon B\u00fchrer","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"B\u00fchrer, Simon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125540278","display_name":"Andreas Plesner","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Plesner, Andreas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125564455","display_name":"Aczel Till","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Till, Aczel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5078339613","display_name":"Roger Wattenhofer","orcid":"https://orcid.org/0000-0002-6339-3134"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wattenhofer, Roger","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.6549000144004822,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.6549000144004822,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.08990000188350677,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.03460000082850456,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/lookup-table","display_name":"Lookup table","score":0.7559000253677368},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6134999990463257},{"id":"https://openalex.org/keywords/modular-design","display_name":"Modular design","score":0.611299991607666},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.6003999710083008},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5723000168800354},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5342000126838684},{"id":"https://openalex.org/keywords/table","display_name":"Table (database)","score":0.5216000080108643},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.45320001244544983}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7738999724388123},{"id":"https://openalex.org/C134835016","wikidata":"https://www.wikidata.org/wiki/Q690265","display_name":"Lookup table","level":2,"score":0.7559000253677368},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6134999990463257},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.611299991607666},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6003999710083008},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5723000168800354},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5342000126838684},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.5216000080108643},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.45320001244544983},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4417000114917755},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4320000112056732},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43140000104904175},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.42289999127388},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4189000129699707},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35420000553131104},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.326200008392334},{"id":"https://openalex.org/C202615002","wikidata":"https://www.wikidata.org/wiki/Q783507","display_name":"Differentiable function","level":2,"score":0.32280001044273376},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.31119999289512634},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3034999966621399},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.2881999909877777},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.271699994802475},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.2614000141620636},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.07400","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.07400","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.07400","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.07400","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.6113669872283936}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,89],"energy":[1],"and":[2,42,85,105,129,137,144],"latency":[3],"costs":[4],"of":[5],"deep":[6],"neural":[7,38,57],"network":[8],"inference":[9,168],"are":[10,40],"increasingly":[11],"driven":[12],"by":[13],"deployment":[14],"rather":[15],"than":[16,159],"training,":[17],"motivating":[18],"hardware-specialized":[19],"alternatives":[20],"to":[21,44,76],"arithmetic-heavy":[22],"models.":[23],"Field-Programmable":[24],"Gate":[25],"Arrays":[26],"(FPGAs)":[27],"provide":[28],"an":[29],"attractive":[30],"substrate":[31],"for":[32,55],"such":[33],"specialization,":[34],"yet":[35],"existing":[36],"FPGA-based":[37],"approaches":[39],"fragmented":[41],"difficult":[43],"compare.":[45],"We":[46],"present":[47],"BitLogic,":[48],"a":[49,92],"fully":[50],"gradient-based,":[51],"end-to-end":[52],"trainable":[53],"framework":[54,90],"FPGA-native":[56],"networks":[58],"built":[59],"around":[60],"Lookup":[61],"Table":[62],"(LUT)":[63],"computation.":[64],"BitLogic":[65],"replaces":[66],"multiply-accumulate":[67],"operations":[68],"with":[69,100,157],"differentiable":[70],"LUT":[71,108,171],"nodes":[72],"that":[73],"map":[74],"directly":[75],"FPGA":[77,148],"primitives,":[78],"enabling":[79],"native":[80],"binary":[81],"computation,":[82],"sparse":[83],"connectivity,":[84],"efficient":[86],"hardware":[87,130,139],"realization.":[88],"offers":[91],"modular":[93],"functional":[94],"API":[95],"supporting":[96],"diverse":[97],"architectures,":[98],"along":[99],"learned":[101],"encoders,":[102],"hardware-aware":[103],"heads,":[104],"multiple":[106],"boundary-consistent":[107],"relaxations.":[109],"An":[110],"automated":[111],"Register":[112],"Transfer":[113],"Level":[114],"(RTL)":[115],"export":[116],"pipeline":[117],"translates":[118],"trained":[119],"PyTorch":[120],"models":[121],"into":[122],"synthesizable":[123],"HDL,":[124],"ensuring":[125],"equivalence":[126],"between":[127],"software":[128],"inference.":[131],"Experiments":[132],"across":[133],"standard":[134],"vision":[135],"benchmarks":[136],"heterogeneous":[138],"platforms":[140],"demonstrate":[141],"competitive":[142],"accuracy":[143,153],"substantial":[145],"gains":[146],"in":[147],"efficiency,":[149],"including":[150],"72.3%":[151],"test":[152],"on":[154],"CIFAR-10":[155],"achieved":[156],"fewer":[158],"0.3M":[160],"logic":[161],"gates,":[162],"while":[163],"attaining":[164],"sub-20":[165],"ns":[166],"single-sample":[167],"using":[169],"only":[170],"resources.":[172]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-02-11T00:00:00"}