[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83036-en":3,"doc-seo-83036-105":29,"detail-sidebar-cat-0-en-105":89},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83036,13056703019404,"Miles","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Determinantal Point Process Sampling for Bioacoustic Active Learning","Eco-acoustic monitoring produces vast audio streams that make annotation costly, motivating active learning to reduce labeling while training accurate biodiversity classifiers. The report introduces CARE-DPP, a batch acquisition method submitted to the BioDCASE Active Learning for Bioacoustics 2026 challenge. CARE-DPP combines class-balanced predictive uncertainty and embedding-space novelty, using a determinantal point process to choose high-quality, non-redundant batches. The uncertainty–novelty balance is annealed across the budget, with an adaptive batch schedule and a mixed candidate pool to stabilize early cycles. Experiments on BirdSet and ATBFL show improved mean development AULC over the baseline.","DETERMINANTAL POINT PROCESS SAMPLING FOR BIOACOUSTIC ACTIVE LEARNING  \nTechnical Report  \nHugo Magaldi 1∗, Gabriel Dubus 1,  \nEco-Anthropologie, Musum National d’Histoire Naturelle, UMR7206, CNRS, Paris, France  \n[hugo.magaldi@mnhn.fr](hugo.magaldi@mnhn.fr)  \narXiv :2607 .06063v 1 [ cs . SD] 7 Jul 2026  \nABSTRACT  \nEco-acoustic monitoring generates vast volumes of audio data, making active learning a promising approach for reducing annotation effort while efficiently training reliable biodiversity classifiers. This report presents CARE-DPP, a batch active-learning acquisition method submitted to BioDCASE Active Learning for Bioacoustics 2026 challenge. The method combines class-balanced predictive uncertainty with embedding-space novelty, while a determinantal point process (DPP) objective selects a high-quality and non-redundant acquisition batch. The uncertainty-novelty balance is annealed over the annotation budget: early cycles emphasize geometric coverage, whereas later cycles increasingly exploit classifier uncertainty. To mitigate unreliable early scores, the DPP candidate pool mixes top-quality candidates with a decreasing proportion of random exploration. An adaptive acquisition schedule uses smaller batches early and larger batches later. Evaluated over five repeats on the BirdSet HSN, POW and UHH subsets and on ATBFL, CARE-DPP obtains a mean development AULC of 0.50 for macro mAP, compared with 0.46 for the official CoreSet baseline. Ablations identify DPP batch diversification and the adaptive acquisition schedule as the largest contributors.  \nIndex Terms— active learning, bioacoustics, determinantal point process, class imbalance, batch diversity  \n1. INTRODUCTION  \nPassive acoustic monitoring can produce far more recordings than experts can annotate. Pool-based active learning aims at addressing this imbalance by repeatedly selecting a small set of unlabeled samples whose annotation is expected to improve a model most efficiently [1, 2] . The BioDCASE Active Learning for Bioacoustics 2026 task standardizes this setting across terrestrial and marine bioacoustics by providing fixed Perch v2 embeddings and a fixed classification pipeline, while restricting system design to the sampling function and acquisition-batch schedule, and ranking methods based on the area under the macro-mAP learning curve (AULC) up to a fixed annotation budget [3, 4] .  \nA successful batch acquisition rule must balance several competing goals. Uncertainty sampling targets decision-boundary examples but can be unreliable when few labels are available and can repeatedly select similar samples. Pure geometric coverage, such as CoreSet [5], is stable early but ignores the evolving classifier and may select acoustically novel samples that are not label-informative. Moreover, macro mAP rewards performance on every class equally, whereas ordinary multilabel uncertainty may be dominated by frequently observed classes.  \n\n| Dataset | Train seg. | Classes | Labels/sample |\n| --- | --- | --- | --- |\n| BirdSet HSN | 6,600 | 19 | 0.524 |\n| BirdSet POW | 2,280 | 41 | 2.833 |\n| BirdSet UHH | 18,319 | 25 | 1.058 |\n| ATBFL (all deployments) | 9,086 | 7 | 2.267 |\n\nTable 1: Development-pool summary. ATBFL statistics aggregate its site-year deployments.  \nWe propose CARE-DPP(Class-balanced Annealed RandomExploration DPP), a method with four components: (i) classbalanced multilabel uncertainty, (ii) cosine novelty relative to the labeled set, (iii) annealed exploration-exploitation weights and candidate-pool exploration, and (iv) DPP-based batch diversification.  \nDeterminantal point processes, originally introduced as repulsive point-process models [6], assign higher probability to subsets whose feature vectors span a large volume [7] . Similarity between vector directions promotes diversity, while externally defined quality scores scale their magnitudes. This makes DPPs well suited to batch active learning, where informative samples should be selected withou","cbCaimI5DklgO0Yh","https://ap.wps.com/l/cbCaimI5DklgO0Yh","pdf",219924,1,4,"English","en",105,"# Abstract\n# Introduction\n# Task and Data\n# Method\n## Class-balanced uncertainty\n## Embedding novelty and annealed quality","[{\"question\":\"What problem does CARE-DPP address in bioacoustic active learning?\",\"answer\":\"Eco-acoustic monitoring generates far more recordings than experts can annotate, so CARE-DPP selects small unlabeled batches expected to improve the classifier efficiently within a fixed annotation budget.\"},{\"question\":\"How does CARE-DPP balance uncertainty, novelty, and diversity when selecting samples?\",\"answer\":\"It combines class-balanced multilabel uncertainty with cosine novelty to labeled embeddings, then uses a determinantal point process objective to diversify the acquisition batch while avoiding redundancy.\"},{\"question\":\"How does the method adapt across the annotation budget during active learning?\",\"answer\":\"The uncertainty–novelty weighting is annealed over cycles: early cycles prioritize geometric coverage, while later cycles increasingly exploit classifier uncertainty. It also uses smaller batches early and larger batches later, with an exploration mix in the candidate pool to mitigate unreliable early scores.\"}]",1784184795,10,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":84,"head_meta":86,"extra_data":88,"updated_unix":27},"determinantal-point-process-sampling-for-bioacoustic-active-learning","",{"@graph":35,"@context":83},[36,52,66],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":21},"https://docshare.wps.com/document/determinantal-point-process-sampling-for-bioacoustic-active-learning/83036/",{"url":51,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":63,"interactionType":64,"userInteractionCount":4},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69,75,79],{"name":70,"@type":71,"acceptedAnswer":72},"What problem does CARE-DPP address in bioacoustic active learning?","Question",{"text":73,"@type":74},"Eco-acoustic monitoring generates far more recordings than experts can annotate, so CARE-DPP selects small unlabeled batches expected to improve the classifier efficiently within a fixed annotation budget.","Answer",{"name":76,"@type":71,"acceptedAnswer":77},"How does CARE-DPP balance uncertainty, novelty, and diversity when selecting samples?",{"text":78,"@type":74},"It combines class-balanced multilabel uncertainty with cosine novelty to labeled embeddings, then uses a determinantal point process objective to diversify the acquisition batch while avoiding redundancy.",{"name":80,"@type":71,"acceptedAnswer":81},"How does the method adapt across the annotation budget during active learning?",{"text":82,"@type":74},"The uncertainty–novelty weighting is annealed over cycles: early cycles prioritize geometric coverage, while later cycles increasingly exploit classifier uncertainty. It also uses smaller batches early and larger batches later, with an exploration mix in the candidate pool to mitigate unreliable early scores.","https://schema.org",{"og:url":51,"og:type":85,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":87,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":90},[91,95,99,103,108,113,118,121,126,129,132],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":92,"show_sort_weight":93,"slug":94},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":96,"show_sort_weight":97,"slug":98},"Literature",80,"literature",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":100,"show_sort_weight":101,"slug":102},"Exam",70,"exam",{"id":104,"doc_module":4,"doc_module_name":45,"category_name":105,"show_sort_weight":106,"slug":107},5,"Comic",60,"comic",{"id":109,"doc_module":4,"doc_module_name":45,"category_name":110,"show_sort_weight":111,"slug":112},6,"Technology",50,"technology",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":130,"show_sort_weight":28,"slug":131},"Lifestyle","lifestyle",{"id":133,"doc_module":4,"doc_module_name":45,"category_name":134,"show_sort_weight":104,"slug":135},19,"General","general"]