{"id":"https://openalex.org/W7162125132","doi":"https://doi.org/10.48550/arxiv.2605.22365","title":"TimeGuard: Channel-wise Pool Training for Backdoor Defense in Time Series Forecasting","display_name":"TimeGuard: Channel-wise Pool Training for Backdoor Defense in Time Series Forecasting","publication_year":2026,"publication_date":"2026-05-21","ids":{"openalex":"https://openalex.org/W7162125132","doi":"https://doi.org/10.48550/arxiv.2605.22365"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.22365","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.22365","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.22365","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136799903","display_name":"Quang Duc Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nguyen, Quang Duc","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136801965","display_name":"Siyuan Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Siyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136731699","display_name":"Yiming Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yiming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136761083","display_name":"Fushuo Huo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huo, Fushuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136768619","display_name":"Dacheng Tao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tao, Dacheng","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/T11326","display_name":"Stock Market Forecasting Methods","score":0.23600000143051147,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11326","display_name":"Stock Market Forecasting Methods","score":0.23600000143051147,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.19200000166893005,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11918","display_name":"Forecasting Techniques and Applications","score":0.0869000032544136,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/backdoor","display_name":"Backdoor","score":0.9976000189781189},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4530999958515167},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.34599998593330383},{"id":"https://openalex.org/keywords/time-series","display_name":"Time series","score":0.2849000096321106},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.2766000032424927}],"concepts":[{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.9976000189781189},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.609499990940094},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4530999958515167},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39100000262260437},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.3483000099658966},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.34599998593330383},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2906000018119812},{"id":"https://openalex.org/C151406439","wikidata":"https://www.wikidata.org/wiki/Q186588","display_name":"Time series","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2718000113964081},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.2630000114440918}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.22365","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.22365","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.22365","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.22365","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/12","score":0.5381421446800232,"display_name":"Responsible consumption and production"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Time":[0],"Series":[1],"Forecasting":[2],"(TSF)":[3],"is":[4],"highly":[5],"vulnerable":[6],"to":[7,16,78,86,124,134],"backdoor":[8,39,100,155],"attacks,":[9],"yet":[10],"effective":[11],"defenses":[12,40,68],"remain":[13],"underexplored":[14],"due":[15],"challenges":[17],"arising":[18],"from":[19],"data":[20,58],"entanglement":[21,59],"and":[22,46,66,73,83,116,142,153],"shifts":[23],"in":[24],"task":[25],"formulation.":[26],"To":[27],"fill":[28],"this":[29],"gap,":[30],"we":[31,96,129],"conduct":[32],"a":[33,98,118],"systematic":[34],"evaluation":[35],"of":[36],"thirteen":[37],"representative":[38],"across":[41,148],"the":[42,113,137,168],"TSF":[43,154],"life":[44],"cycle":[45],"analyze":[47],"their":[48],"failure":[49],"modes.":[50],"Our":[51,106],"results":[52],"reveal":[53],"two":[54],"fundamental":[55],"issues:":[56],"(1)":[57],"induces":[60],"channel-level":[61],"signal":[62,126],"dilution,":[63],"rendering":[64],"sample-filtering":[65],"trigger-synthesis":[67],"ineffective":[69],"at":[70,89],"localizing":[71],"backdoors;":[72],"(2)":[74],"task-formulation":[75],"shift":[76],"leads":[77],"training-loss":[79],"degeneration,":[80],"causing":[81],"poisoned":[82],"clean":[84,173],"windows":[85],"become":[87],"indistinguishable":[88],"training":[90,111,141],"stages.":[91],"Based":[92],"on":[93],"these":[94],"findings,":[95],"propose":[97],"training-time":[99],"defense":[101],"for":[102],"TSF,":[103],"termed":[104],"TimeGuard.":[105],"method":[107],"adopts":[108],"channel-wise":[109],"pool":[110,120,139],"as":[112],"core":[114],"paradigm":[115],"initializes":[117],"high-confidence":[119],"using":[121],"time-aware":[122],"criteria":[123],"mitigate":[125],"dilution.":[127],"Moreover,":[128],"introduce":[130],"distance-regularized":[131],"loss":[132,144],"selection":[133],"progressively":[135],"expand":[136],"reliable":[138],"during":[140],"ease":[143],"degeneration.":[145],"Extensive":[146],"experiments":[147],"multiple":[149],"datasets,":[150],"forecasting":[151],"architectures,":[152],"attacks":[156],"demonstrate":[157],"that":[158],"TimeGuard":[159],"substantially":[160],"improves":[161],"robustness,":[162],"boosting":[163],"$\\mathrm{MAE}_\\mathrm{P}$":[164],"by":[165],"$1.96\\times$":[166],"over":[167],"leading":[169],"baseline,":[170],"while":[171],"preserving":[172],"performance":[174],"within":[175],"5%":[176],"$\\mathrm{MAE}_\\mathrm{C}$.":[177]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-23T00:00:00"}