[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-116939-en":3,"doc-seo-116939-105":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},116939,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Out-of-distribution generalisation in machine learning - Doctoral dissertation","Machine learning performs reliably when training and test data follow the same distribution, yet it often fails under distribution shift due to the non-stationary and heterogeneous nature of real-world data. The dissertation develops out-of-distribution generalisation by presenting three studies focused on new tools, including algorithms, theoretical results, experimental results, and datasets. A unifying driver is modularity, aiming to prevent rigid memorisation and promote flexible generalisation. Contributions cover learning from multiple training distributions, improved supervised image recognition via multi-level feature aggregation and augmentation, and multi-agent communication through graph referential games, including realistic population-based communication.","Out-of-distribution generalisation in  \nmachine learning  \nAgnieszka Słowik  \nLucy Cavendish College  \nThis dissertation is submitted for the degree of Doctor of Philosophy.  \nDecember 2022  \nTo my mum Izabela.  \nDeclaration  \nThis dissertation is the result of my own work and includes nothing which is the outcome of work done in collaboration except as declared in the Preface and specified in the text. It is not substantially the same as any that I have submitted, or am concurrently submitting, for a degree or diploma or other qualification at the University of Cambridge or any other University or similar institution except as declared in the Preface and specified in the text. I further state that no substantial part of my dissertation has already been submitted, or is being concurrently submitted, for any such degree, diploma or other qualification at the University of Cambridge or any other University or similar institution except as declared in the Preface and specified in the text. This dissertation does not exceed the prescribed limit of 60 000 words.  \nAbstract  \nOut-of-distribution generalisation in machine learning  \nAgnieszka Słowik  \nMachine learning has proven extremely useful in many applications in recent years. However, a lot of these success stories stem from evaluating the algorithms on data very similar to that they were trained on. When applied to a new data distribution, machine learning algorithms have been shown to fail. Given the non-stationary and heterogeneous nature of real-world data, a better grasp of out-of-distribution generalisation is needed for algorithms to be widely deployed and trusted.  \nMy thesis presents three research studies that aim to investigate and develop the field of outof-distribution generalisation. The central goal of these research efforts is to produce new tools, such as algorithms, theoretical results, experimental results and datasets, to improve understanding and performance of machine learning methods in the face of distribution shift. The high-level idea that drives these research efforts across three machine learning scenarios is modularity – the quality of consisting of separate parts that form a whole when combined. Modular approaches are hypothesised to steer the machine learning methods away from rigid memorisation of examples and towards more flexible and ‘more intelligent’learning that supports generalisation.  \nIn my first contribution, I approach the thesis goal from the perspective of learning from multiple training distributions. The contribution to this line of research is twofold. First, I present a new standardised suite of tasks for evaluation and comparison of out-ofdistribution generalisation algorithms. Second, I state a set of new theoretical results that fill an existing gap between data-centric and algorithmic approaches to out-of-distribution generalisation. These theoretical findings guide a new set of practical recommendations on how to employ the algorithmic approach.  \nIn the second contribution, I tackle generalisation in the common learning setup of supervised image recognition. In this context, I first investigate the effect of multi-level feature aggregation on generalisation, and demonstrate that augmentation with one of the considered methods consistently improves the performance. Second, I propose a set of  \nsimple image datasets that can be used as a stepping stone for evaluation and comparison of image classification methods in terms of out-of-distribution generalisation.  \nFinally, I delve into the learning scenarios where multiple neural networks communicate to solve a shared task. This work supports the thesis goal in two ways. First, I propose anew environment, graph referential games, and present results on the influence of data representation and the corresponding data representation learning methods on out-ofdistribution generalisation. These results connect the previously disjoint fields of graph representation learning and emer","cbCairANaQHmgwl1","https://ap.wps.com/l/cbCairANaQHmgwl1","pdf",7764856,1,214,"English","en",105,"# Abstract\n## Research goal and motivation\n## Contribution 1: learning from multiple training distributions\n## Contribution 2: supervised image recognition\n## Contribution 3: multi-agent communication","[{\"question\":\"Why is out-of-distribution generalisation necessary in machine learning?\",\"answer\":\"Many successes come from evaluating on data similar to training. Real-world data is non-stationary and heterogeneous, so algorithms can fail under distribution shift without robust out-of-distribution generalisation.\"},{\"question\":\"What unifying idea guides the dissertation’s three research studies?\",\"answer\":\"Modularity: separating parts that combine into a whole. Modular approaches aim to steer models away from rigid memorisation toward more flexible learning that supports generalisation.\"},{\"question\":\"What are the main contributions across the three research scenarios?\",\"answer\":\"They include: (1) a standardised task suite and theoretical results for learning from multiple training distributions; (2) analysis of multi-level feature aggregation in supervised image recognition plus improved performance via augmentation and new stepping-stone datasets; (3) graph referential games and population-based communication grounded in realistic images, linking graph representation learning with emergent communication.\"}]","Out-of-distribution generalisation in machine learning - Doctoral dissertation | PDF",1785672644,539,{"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},"out-of-distribution-generalisation-in-machine-learning-doctoral-dissertation","",{"@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/out-of-distribution-generalisation-in-machine-learning-doctoral-dissertation/116939/",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},"Why is out-of-distribution generalisation necessary in machine learning?","Question",{"text":75,"@type":76},"Many successes come from evaluating on data similar to training. Real-world data is non-stationary and heterogeneous, so algorithms can fail under distribution shift without robust out-of-distribution generalisation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What unifying idea guides the dissertation’s three research studies?",{"text":80,"@type":76},"Modularity: separating parts that combine into a whole. Modular approaches aim to steer models away from rigid memorisation toward more flexible learning that supports generalisation.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the main contributions across the three research scenarios?",{"text":84,"@type":76},"They include: (1) a standardised task suite and theoretical results for learning from multiple training distributions; (2) analysis of multi-level feature aggregation in supervised image recognition plus improved performance via augmentation and new stepping-stone datasets; (3) graph referential games and population-based communication grounded in realistic images, linking graph representation learning with emergent communication.","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"]