[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122007-en":3,"doc-seo-122007-105":30,"detail-sidebar-cat-0-en-105":91},{"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":27,"seo_description":14,"update_tm":28,"read_time":29},122007,5909877438554,"Maeve","https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272",8,"Research & Report","Moderately Distributional Exploration for Domain Generalization - Abstract and Introduction","Domain generalization (DG) tackles distribution shift between training domains and unseen target domains. While generating new domains can help, performance depends on how close generated distributions are to target ones. The paper argues that overly large uncertainty sets in distributionally robust optimization can reduce prediction confidence by including semantically unrelated domains. It introduces moderately distributional exploration (MODE), exploring only an uncertainty subset sharing semantic factors with training domains, yielding provable generalization and competitive experimental results.","Moderately Distributional Exploration for Domain Generalization  \nRui Dai 1 Yonggang Zhang 2 † Zhen Fang 3 Bo Han 2 Xinmei Tian 1 4 †  \nAbstract  \nDomain generalization (DG) aims to tackle the distribution shift between training domains and unknown target domains. Generating new domains is one of the most effective approaches, yet its performance gain depends on the distribution discrepancy between the generated and target domains. Distributionally robust optimization is promising to tackle distribution discrepancy by exploring domains in an uncertainty set.  \nHowever, the uncertainty set may be overwhelmingly large, leading to low-confidence prediction in DG. It is because a large uncertainty set could introduce domains containing semantically different factors from training domains. To address this issue, we propose to perform a moderately distributional exploration (MODE) for domain generalization. Specifically, MODE performs distribution exploration in an uncertainty subset that shares the same semantic factors with the training domains. We show that MODE can endow models with provable generalization performance on unknown target domains. The experimental results show that MODE achieves competitive performance compared to state-of-the-art baselines.  \n1. Introduction  \nDeep neural networks (DNNs) have achieved exciting performance on various tasks. The successes of DNNs heavily depend on an underlying assumption that the training domains and target domain share the same distribution. However, this assumption may not hold in some practical sce-  \n1University of Science and Technology of China, Hefei, China 2Department of Computer Science, Hong Kong Baptist University, HongKong, China 3Australian Artificial Intelligence Institute, University of Technology Sydney, Sydney, Australia 4Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China. Correspondence to: Xinmei Tian \u003C[xinmei@ustc.edu.cn](xinmei@ustc.edu.cn) >, Yonggang Zhang \u003C[csygzhang@comp.hkbu.edu.hk](csygzhang@comp.hkbu.edu.hk)>.  \nProceedings of the 40 th International Conference on Machine Learning, Honolulu, Hawaii, USA. PMLR 202, 2023 . Copyright 2023 by the author(s) .  \nnarios, which leads to the failure of DNNs. To release this assumption, researchers have studied a more practical learning setting called Domain Generalization (DG) (Muandet et al., 2013 ; Ye et al., 2021 ; Shen et al., 2021) . The goal of DG is to train models using training domains such that these models can generalize well in the unknown target domain which shares the same semantics with the training domains.  \nTo generalize well on the unknown target domains, previous works introduce a domain generation strategy, enhancing the performance of DNNs by generating new domains (Zhou et al., 2020b ;a; Wang et al., 2021 ; Xu et al., 2021) . The underlying intuition of this approach is that learning with many generated domains could make DNNs robust against domain shifts. However, it remains challenging how to construct new domains to achieve a provable generalization performance on target domains. Namely, it is challenging to guarantee a mitigated distribution discrepancy between the generated domains and target domains. Accordingly, the generated domains may fail to promote generalizability or even cause performance degradation of DNNs. The reason lies in the fact that target domains are unknown in the training process, leading to an uncontrollable distribution discrepancy between the generated and the target domains.  \nDistributionally Robust Optimization (DRO) is a possible strategy to tackle the distribution discrepancy between training and target domains (Csiszar, 1967 ; Namkoong & Duchi, 2016 ; Staib & Jegelka, 2019) . The intuition of DRO is to extend one distribution to a distribution space, i.e., uncertainty set, and uses the worst-case distribution in the uncertainty set for model training (Sinha et al., 2018 ; Michel et al., 2021 ; Mehra et al., 2022","cbCaiiS5Z02tWbBq","https://ap.wps.com/l/cbCaiiS5Z02tWbBq","pdf",10374882,1,32,"English","en",105,"# Abstract\n# 1. Introduction\n## Domain generalization motivation and challenge\n## Distributionally Robust Optimization and low-confidence issue\n## MODE: moderately distributional exploration idea\n## Causal factorization assumption","[{\"question\":\"What problem does domain generalization aim to solve?\",\"answer\":\"It addresses the distribution shift between training domains and unknown target domains, while preserving the shared semantics between them.\"},{\"question\":\"Why can directly applying DRO to DG lead to limited improvements?\",\"answer\":\"Large uncertainty sets may include semantically inconsistent domains, causing low-confidence predictions and reducing practical performance gains.\"},{\"question\":\"How does MODE improve over exploring an entire uncertainty set?\",\"answer\":\"MODE explores a smaller uncertainty subset that shares the same semantic factors with training domains, which mitigates the low-confidence issue and supports provable generalization.\"}]","Moderately Distributional Exploration for Domain Generalization - Abstract and Introduction | PDF",1785808264,81,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"moderately-distributional-exploration-for-domain-generalization-abstract-and-introduction","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/moderately-distributional-exploration-for-domain-generalization-abstract-and-introduction/122007/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does domain generalization aim to solve?","Question",{"text":75,"@type":76},"It addresses the distribution shift between training domains and unknown target domains, while preserving the shared semantics between them.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why can directly applying DRO to DG lead to limited improvements?",{"text":80,"@type":76},"Large uncertainty sets may include semantically inconsistent domains, causing low-confidence predictions and reducing practical performance gains.",{"name":82,"@type":73,"acceptedAnswer":83},"How does MODE improve over exploring an entire uncertainty set?",{"text":84,"@type":76},"MODE explores a smaller uncertainty subset that shares the same semantic factors with training domains, which mitigates the low-confidence issue and supports provable generalization.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]