{"id":"https://openalex.org/W4319870541","doi":"https://doi.org/10.1145/3543622.3573044","title":"Accelerating Neural-ODE Inference on FPGAs with Two-Stage Structured Pruning and History-based Stepsize Search","display_name":"Accelerating Neural-ODE Inference on FPGAs with Two-Stage Structured Pruning and History-based Stepsize Search","publication_year":2023,"publication_date":"2023-02-10","ids":{"openalex":"https://openalex.org/W4319870541","doi":"https://doi.org/10.1145/3543622.3573044"},"language":"en","primary_location":{"id":"doi:10.1145/3543622.3573044","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3543622.3573044","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2023 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5043971987","display_name":"Lei Cai","orcid":"https://orcid.org/0000-0001-5637-865X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Lei Cai","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100610519","display_name":"Jing Wang","orcid":"https://orcid.org/0000-0003-4146-2749"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jing Wang","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021889090","display_name":"Lianfeng Yu","orcid":"https://orcid.org/0000-0001-8113-2600"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lianfeng Yu","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043075091","display_name":"Bonan Yan","orcid":"https://orcid.org/0000-0002-3052-9330"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bonan Yan","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072208883","display_name":"Yaoyu Tao","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yaoyu Tao","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5057584787","display_name":"Yuchao Yang","orcid":"https://orcid.org/0000-0003-4674-4059"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuchao Yang","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5043971987"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":1.4803,"has_fulltext":false,"cited_by_count":9,"citation_normalized_percentile":{"value":0.80051,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"177","last_page":"183"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11206","display_name":"Model Reduction and Neural Networks","score":0.9994000196456909,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11206","display_name":"Model Reduction and Neural Networks","score":0.9994000196456909,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12810","display_name":"Real-time simulation and control systems","score":0.9807000160217285,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10558","display_name":"Advancements in Semiconductor Devices and Circuit Design","score":0.9761999845504761,"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/speedup","display_name":"Speedup","score":0.9180431365966797},{"id":"https://openalex.org/keywords/ode","display_name":"Ode","score":0.7790871858596802},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.755457878112793},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6179517507553101},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6060572862625122},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5706986784934998},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5095108151435852},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.4856528043746948},{"id":"https://openalex.org/keywords/ordinary-differential-equation","display_name":"Ordinary differential equation","score":0.47186702489852905},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4418465793132782},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.34375080466270447},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2708793878555298},{"id":"https://openalex.org/keywords/differential-equation","display_name":"Differential equation","score":0.2336723506450653},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.17820286750793457},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.15147095918655396},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.10716471076011658}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.9180431365966797},{"id":"https://openalex.org/C34862557","wikidata":"https://www.wikidata.org/wiki/Q178985","display_name":"Ode","level":2,"score":0.7790871858596802},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.755457878112793},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6179517507553101},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6060572862625122},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5706986784934998},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5095108151435852},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.4856528043746948},{"id":"https://openalex.org/C51544822","wikidata":"https://www.wikidata.org/wiki/Q465274","display_name":"Ordinary differential equation","level":3,"score":0.47186702489852905},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4418465793132782},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.34375080466270447},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2708793878555298},{"id":"https://openalex.org/C78045399","wikidata":"https://www.wikidata.org/wiki/Q11214","display_name":"Differential equation","level":2,"score":0.2336723506450653},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.17820286750793457},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15147095918655396},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.10716471076011658},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3543622.3573044","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3543622.3573044","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2023 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W424133987","https://openalex.org/W768639568","https://openalex.org/W1998583908","https://openalex.org/W2194775991","https://openalex.org/W2531409750","https://openalex.org/W2944423816","https://openalex.org/W3131639738","https://openalex.org/W3133438734","https://openalex.org/W3138516171","https://openalex.org/W3159008243","https://openalex.org/W3206719067","https://openalex.org/W4211209812","https://openalex.org/W4212774754","https://openalex.org/W4242177601","https://openalex.org/W4287906222","https://openalex.org/W4300831640"],"related_works":["https://openalex.org/W3082668976","https://openalex.org/W2517973022","https://openalex.org/W1543121148","https://openalex.org/W1580703421","https://openalex.org/W2618518959","https://openalex.org/W2231364979","https://openalex.org/W4295745414","https://openalex.org/W4382202915","https://openalex.org/W2489516236","https://openalex.org/W4297273858"],"abstract_inverted_index":{"Neural":[0],"ordinary":[1],"differential":[2],"equation":[3],"(Neural-ODE)":[4],"outperforms":[5],"conventional":[6],"deep":[7],"neural":[8],"networks":[9],"(DNNs)":[10],"in":[11,40,90,117,159],"modeling":[12],"continuous-time":[13],"or":[14],"dynamical":[15],"systems":[16],"by":[17,86,113],"adopting":[18],"numerical":[19,41],"ODE":[20],"integration":[21,56,103],"onto":[22],"a":[23,71,95,141,163,171,179,204],"shallow":[24],"embedded":[25,92,138],"NN.":[26],"However,":[27],"Neural-ODE":[28,69,194],"suffers":[29],"from":[30],"slow":[31],"inference":[32,212],"due":[33],"to":[34,66,115,135,155,191,203,217],"the":[35,83,91,107,175,192,198,211],"costly":[36],"iterative":[37],"stepsize":[38,97,112],"search":[39,98],"integration,":[42],"especially":[43],"when":[44],"using":[45,178,197],"higher-order":[46,144],"Runge-Kutta":[47],"(RK)":[48],"methods":[49],"and":[50,94,140,168],"smaller":[51],"error":[52],"tolerance":[53],"for":[54,109],"improved":[55],"accuracy.":[57],"In":[58],"this":[59],"work,":[60],"we":[61],"first":[62],"present":[63],"algorithmic":[64],"techniques":[65],"speedup":[67,129,189,213],"RK-based":[68],"inference:":[70],"two-stage":[72],"coarse-grained/fine-grained":[73],"structured":[74,157],"pruning":[75],"method":[76,99,184],"based":[77,100,126],"on":[78,101,127,162,170,207],"top-K":[79],"sparsification":[80],"that":[81,105],"reduces":[82,106],"overall":[84],"computations":[85],"more":[87,180,186],"than":[88,187],"60%":[89],"NN":[93,139],"history-based":[96],"past":[102],"steps":[104],"latency":[108],"reaching":[110],"accepted":[111],"up":[114,216],"77%":[116],"RK":[118,145,183],"methods.":[119,146],"A":[120],"reconfigurable":[121],"hardware":[122],"architecture":[123],"is":[124,153],"co-designed":[125],"proposed":[128],"techniques,":[130],"featuring":[131],"three":[132],"processing":[133],"loops":[134],"support":[136],"programmable":[137],"variety":[142,172],"of":[143,173],"Sparse":[147],"activation":[148],"processor":[149],"with":[150],"multi-dimensional":[151],"sorters":[152],"designed":[154],"exploit":[156],"sparsity":[158],"activations.":[160],"Implemented":[161],"Xilinx":[164],"Virtex-7":[165],"XC7VX690T":[166],"FPGA":[167,195],"experimented":[169],"datasets,":[174],"prototype":[176],"accelerator":[177,196],"complex":[181],"3rd-order":[182],"achieves":[185],"2.6x":[188],"compared":[190],"latest":[193],"simplest":[199],"Euler":[200],"method.":[201],"Compared":[202],"software":[205],"execution":[206],"Nvidia":[208],"A100":[209],"GPU,":[210],"can":[214],"be":[215],"18x.":[218]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}