[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117539-en":3,"doc-seo-117539-105":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},117539,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Continual Learning: Applications and the Road Forward","Continual learning enables machine learning models to learn continuously from new data while retaining previously acquired knowledge without catastrophic forgetting. The work reframes the field by asking why continual learning matters and surveys recent papers from major conferences, showing memory-constrained settings dominate. It identifies five open problems—model editing, personalization and specialization, on-device learning, faster (re-)training, and reinforcement learning—and argues continual learning is central to their solutions. A comparison between desiderata and current assumptions leads to four future research directions.","Continual Learning: Applications and the Road Forward  \nEli Verwimp∗  \nRahaf Aljundi  \nShai Ben-David Matthias Bethge Andrea Cossu Alexander Gepperth Tyler L. Hayes  \nEyke Hüllermeier Christopher Kanan Dhireesha Kudithipudi Christoph H. Lampert Martin Mundt Razvan Pascanu Adrian Popescu Andreas S. Tolias Joost van de Weijer Bing Liu  \nVincenzo Lomonaco Tinne Tuytelaars Gido M. van de Ven  \nKU Leuven, Belgium  \nToyota Motor Europe, Belgium  \nUniversity of Waterloo, and Vector Institute, Ontario, Canada  \nUniversity of Tübingen, Germany University of Pisa, Italy  \nUniversity of Applied Sciences Fulda, Germany NAVER LABS Europe, France  \nUniversity of Munich (LMU), Germany University of Rochester, Rochester, NY, USA University of Texas at San Antonio, TX, USA Institute of Science and Technology Austria (ISTA)  \nTU Darmstadt & hessian.AI, Germany Google DeepMind, UK  \nUniversité Paris-Saclay, CEA, LIST, France Baylor College of Medicine, Houston, TX, USA Computer Vision Center, UAB, Barcelona, Spain University of Illinois at Chicago, USA  \nUniversity of Pisa, Italy KU Leuven, Belgium  \nKU Leuven, Belgium  \nReviewed on OpenReview: [https://openreview.net/forum?id=axBIMcGZn9](https://openreview.net/forum?id=axBIMcGZn9)  \nAbstract  \nContinual learning is a subfield of machine learning, which aims to allow machine learning models to continuously learn on new data, by accumulating knowledge without forgetting what was learned in the past. In this work, we take a step back, and ask: “ Why should one care about continual learning in the first place?”. We set the stage by examining recent continual learning papers published at four major machine learning conferences, and show that memory-constrained settings dominate the field. Then, we discuss five open problems in machine learning, and even though they might seem unrelated to continual learning at first sight, we show that continual learning will inevitably be part of their solution. These problems are model editing, personalization and specialization, on-device learning, faster (re-)training and reinforcement learning. Finally, by comparing the desiderata from these unsolved problems and the current assumptions in continual learning, we highlight and discuss four future directions for continual learning research. We hope that this work offers an interesting perspective on the future of continual learning, while displaying its potential value and the paths we have to pursue in order to make it successful. This work is the result of the many discussions the authors had at the Dagstuhl seminar on Deep Continual Learning, in March 2023 .  \n∗ Corresponding author: [eli.verwimp@kuleuven.be](eli.verwimp@kuleuven.be)  \n1 Introduction  \nContinual learning, sometimes referred to as lifelong learning or incremental learning, is a subfield of machine learning that focuses on the challenging problem of incrementally training models on a stream of data with the aim of accumulating knowledge over time. This setting calls for algorithms that can learn new skills with minimal forgetting of what they had learned previously, transfer knowledge across tasks, and smoothly adapt to new circumstances when needed. This is in contrast with the traditional setting of machine learning, which typically builds on the premise that all data, both for training and testing, are sampled i.i.d. (independent and identically distributed) from a single, stationary data distribution.  \nDeep learning models in particular are in need of continual learning capabilities. A first reason for this is their strong dependence on data. When trained on a stream of data whose underlying distribution changes over time, deep learning models tend to adapt to the most recent data, thereby “catastrophically” forgetting the information that had been learned earlier (French, 1999) . Secondly, continual learning capabilities could reduce the very long training times of deep learning models. When new data are available, current industry practic","cbCaikraQpz9mQrW","https://ap.wps.com/l/cbCaikraQpz9mQrW","pdf",1367966,1,21,"English","en",105,"# Introduction\n## Learning from a data stream\n## Need for continual learning in deep learning\n## Naive approaches and their limitations\n## Memory-constrained trade-offs","[{\"question\":\"What problem does continual learning address?\",\"answer\":\"Continual learning incrementally trains models on a stream of data to accumulate knowledge over time while minimizing forgetting of earlier information.\"},{\"question\":\"Why are deep learning models especially in need of continual learning?\",\"answer\":\"Because changing data distributions can cause catastrophic forgetting, and because fully retraining on all past and new data is computationally costly and often unsustainable.\"},{\"question\":\"Which open problems are discussed as inevitably linked to continual learning?\",\"answer\":\"The paper discusses model editing, personalization and specialization, on-device learning, faster (re-)training, and reinforcement learning.\"}]","Continual Learning: Applications and the Road Forward | PDF",1785676798,53,{"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},"continual-learning-applications-and-the-road-forward","",{"@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/continual-learning-applications-and-the-road-forward/117539/",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","2026-08-02",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 continual learning address?","Question",{"text":75,"@type":76},"Continual learning incrementally trains models on a stream of data to accumulate knowledge over time while minimizing forgetting of earlier information.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why are deep learning models especially in need of continual learning?",{"text":80,"@type":76},"Because changing data distributions can cause catastrophic forgetting, and because fully retraining on all past and new data is computationally costly and often unsustainable.",{"name":82,"@type":73,"acceptedAnswer":83},"Which open problems are discussed as inevitably linked to continual learning?",{"text":84,"@type":76},"The paper discusses model editing, personalization and specialization, on-device learning, faster (re-)training, and reinforcement learning.","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"]