[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-116980-en":3,"doc-seo-116980-105":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},116980,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Machine learning, justification, and computational reliabilism","Reliable machine learning concerns epistemic justification: how a system can justify believing its own outputs. Because many ML methods are opaque and validation may be difficult, standard approaches to reliability such as transparency require exposing internal mechanisms and algorithmic properties. This paper develops computational reliabilism (CR) as an ML-centered reliabilist framework. CR grounds justification in external reliability indicators—e.g., validation procedures and knowledge-based integration—without requiring access to inner workings.","Machine learning, justification, and computational  \nreliabilism  \nJuan M. Dur´an  \nUnpublished manuscript  \nAbstract  \nThis article asks the question,“what is reliable machine learning?” As I intend to answer it, this is a question about epistemic justification. Reliable machine learning gives justification for believing its output. Current approaches to reliability (e.g. , transparency) involve showing the inner workings of an algorithm (functions, variables, etc.) and how they render outputs. We then have justification for believing the output because we know how it was computed. Thus, justification is contingent on what can be shown about the algorithm, its properties, and its behavior. In this paper, I defend computational reliabilism (CR) . CR is a computationally-inspired off-shoot of process reliabilism that does not require showing the inner workings of an algorithm. CR credits reliability to machine learning by identifying reliability indicators external to the algorithm (validation methods, knowledge-based integration, etc.) . Thus, we have justification for believing the output of machine learning when we have identified the appropriate reliability indicators. CR is advanced as amore suitable epistemology for machine learning. The main goal of this article is to lay the groundwork for CR, how it works, and what we can expect as a justificatory framework for reliable machine learning.  \n1 Introduction  \nThe use of Machine Learning (ML) for scientific purposes is delivering remarkable results. A couple of examples will suffice to show this. In molecular biology, AlphaFold can predict protein structures with atomic accuracy for cases in which no similar structure is known (Jumper et al., 2021) . In medicine, BenevolentAI has combined structured with unstructured biomedical data sources to identify rheumatoid arthritis drugs like baricitinib as therapeutics for COVID-19 symptoms (Medeiros, 2021) . It is clear that ML  \ncan successfully extend the class of tractable chemistry, biology, physics, and medicine, broadening the range of modeling and experimental capabilities of researchers.  \nYet, unlike many other methods, ML’s scientific value cannot be easily determined by association with a body of scientific knowledge, by means of adequacy to empirical data, or supported by theoretical constructs (e.g., explanation and observation) . This for a variety of reasons. ML is epistemically and methodologically opaque (Humphreys, 2009), making it difficult to associate a given algorithm and its output with the general scientific canon; and empirical phenomena are often temporarily, spatially, or cognitively inaccessible for empirical validation of the model, casting doubts over the representational value of these systems. As a consequence, there are significant impediments for making claims about our reliance on these systems and their output.  \nWhen confronted with these issues, philosophers and computer scientists gravitate towards transparency. Transparency is an umbrella term capturing diverse methods linking the internal mechanisms and properties of algorithms to its outputs (Creel, 2020; Wachter et al., 2018; Ribeiro et al., 2016) . To see how transparency works, consider BenevolentAI. At its core, BenevolentAI is a search engine combining structured with unstructured biomedical data sources, drug industry data, and automated retrieval of information from scientific research papers. The data is curated and standardized via data analysis and data fabric. It is then fed into knowledge graphs that structure the data into relationships between diseases, genes, and different drugs (Smith et al., 2021) . Richardson led the team that used BenevolentAI to identify rheumatoid arthritis drugs – notably baricitinib – as suitable therapeutics for COVID-19 symptoms (Richardson et al., 2020) . In order to justify Richardson’s reliance on this output, partisans of transparency make efforts to show how baricitinib obtains from procedure","cbCaihYl9tnGbbPB","https://ap.wps.com/l/cbCaihYl9tnGbbPB","pdf",256213,1,21,"English","en",105,"# Abstract\n# Introduction","[{\"question\":\"What problem does the paper address about reliable machine learning?\",\"answer\":\"It asks what it means for machine learning to be reliable, framing reliability as a question of epistemic justification for believing ML outputs.\"},{\"question\":\"How do current reliability approaches typically justify ML outputs?\",\"answer\":\"They often rely on transparency by revealing an algorithm’s internal mechanisms and properties, so justification follows from knowing how the output was computed.\"},{\"question\":\"What is computational reliabilism (CR), and how does it differ from process reliabilism?\",\"answer\":\"CR credits ML reliability through external reliability indicators, such as validation methods and knowledge-based integration, rather than requiring transparency of inner workings. It also does not assume in advance that ML is a reliable belief-forming method.\"}]","Machine learning, justification, and computational reliabilism | PDF",1785672947,53,{"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},"machine-learning-justification-and-computational-reliabilism","",{"@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/machine-learning-justification-and-computational-reliabilism/116980/",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 the paper address about reliable machine learning?","Question",{"text":75,"@type":76},"It asks what it means for machine learning to be reliable, framing reliability as a question of epistemic justification for believing ML outputs.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do current reliability approaches typically justify ML outputs?",{"text":80,"@type":76},"They often rely on transparency by revealing an algorithm’s internal mechanisms and properties, so justification follows from knowing how the output was computed.",{"name":82,"@type":73,"acceptedAnswer":83},"What is computational reliabilism (CR), and how does it differ from process reliabilism?",{"text":84,"@type":76},"CR credits ML reliability through external reliability indicators, such as validation methods and knowledge-based integration, rather than requiring transparency of inner workings. It also does not assume in advance that ML is a reliable belief-forming method.","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"]