[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117974-en":3,"doc-seo-117974-105":29,"detail-sidebar-cat-0-en-105":90},{"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":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},117974,3848291630094,"Emma Wilson","https://eur-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","Bridging Machine Learning and Sciences - Opportunities and Challenges","Machine learning in scientific research has accelerated with major advances, especially anomaly detection enhanced by deep neural networks for out-of-distribution detection on high-dimensional data. The work critically examines practical prospects for applying these methods across scientific domains, focusing on data universality, experimental protocol design, and model robustness. Case examples highlight how transferable practices and domain-specific obstacles can coexist, supporting the emergence of a new interdisciplinary research paradigm.","Bridging Machine Learning and Sciences: Opportunities and Challenges  \nTaoli Cheng  \nMila, University of Montreal [chengtaoli.1990@gmail.com](chengtaoli.1990@gmail.com)  \narXiv :2210 . 13441v2 [ stat .ML] 2 Nov 2023  \nAbstract  \nThe application of machine learning in sciences has seen exciting advances in recent years. Asa widely applicable technique, anomaly detection has been long studied in the machine learning community. Especially, deep neural netsbased out-of-distribution detection has made great progress for high-dimensional data. Recently, these techniques have been showing their potential in scientific disciplines. We take a critical look at their applicative prospects including data universality, experimental protocols, model robustness, etc. We discuss examples that display transferable practices and domain-specific challenges simultaneously, providing a starting point for establishing a novel interdisciplinary research paradigm in the near future.  \n1. Introduction  \nThe advances in the deep learning revolution have been expanding their influence in many domains and accelerating research in a generalized interdisciplinary manner. Neural nets serving as general function approximators have been employed in scientific applications including object identification/classification, anomaly/novelty detection, autonomous control, and neural net-based simulation, etc. Great successes have been made in multiple scientific disciplines. A typical example is AlphaFold (Jumper et al., 2021) for accurate protein structure prediction. At the same time, alot of progress has been made in physical sciences (Baldiet al., 2014 ; Ribli et al., 2019), biology (Ching et al., 2018), molecule generation / drug discovery (Gottipati et al., 2020), medical imaging (Zhang et al., 2020), etc.  \nDespite the successes, there are many challenges or unrecognized pitfalls in transporting machine learning techniques into more traditional science domains. Not being aware of these possible pitfalls could result in vain efforts and sometimes catastrophic consequences in real-world model deployment. In the following, we take a holistic look at the current collaborative scheme in machine learning applications for sciences in this new cross-disciplinary research era.  \nThe differences between general machine learning (mainly focused on computer vision (CV) and natural language processing (NLP)) and tailored scientific applications reside in all parts of the pipeline. The following aspects build an intertwined picture in the modern machine learning-assisted scientific discovery:  \n• Nature of the data: In natural sciences especially physical sciences, scientists usually strictly design (and usually simplify) the experimental environments to probe the considered phenomena. Thus the nature of scientific data is under control to be free of external noises due to the laboratory settings. (And normally systematic uncertainties can be well estimated through control datasets) However, the format of the data can be more complex (and non-human-readable) compared with natural images.  \n• Inference process: Model inference in real-world settings can come in complicated and varying formats. As for scientific applications, usually, the experimental focus defines the inference process. And consequently, the inference process affects the result interpretation.  \n• Benchmarks: In the machine learning community usually benchmark datasets used for model evaluation is restricted to a few public datasets such as MNIST (Lecun et al., 1998), ImageNet (Deng et al., 2009), CIFAR (Krizhevsky, 2009), or SVHN (Netzer et al., 2011) . This inevitably results in “overfitted” strategies and research focuses. In contrast, domain sciences haven’t yet built common datasets for model training and evaluation, sometimes resulting in difficulties in model comparison.  \n• Uncertainty quantification: Uncertainty of deep neural net outputs can be hard to quantify due to the complexity involved. When","cbCaisR5AcmjVkA6","https://ap.wps.com/l/cbCaisR5AcmjVkA6","pdf",526407,1,"English","en",105,"# Introduction\n## Pipeline Differences Between General ML and Scientific Applications\n# Scientific Discovery","[{\"question\":\"Why is anomaly detection especially important for bridging machine learning and scientific applications?\",\"answer\":\"Anomaly detection has long been studied, and deep neural networks have improved out-of-distribution detection for complex high-dimensional data. This capability is increasingly relevant in scientific settings.\"},{\"question\":\"What key differences separate general machine learning from tailored scientific applications?\",\"answer\":\"Differences span the nature of data, the inference process driven by experimental focus, benchmark limitations, uncertainty quantification needs, and challenges in generalization and robustness under simulation-to-real shifts.\"},{\"question\":\"How does the document suggest improving model transfer from simulation to real scientific environments?\",\"answer\":\"It emphasizes adapting the workflow to the needs of the scientific domain, including accounting for dataset shift and precision requirements when transitioning from simulation data with labels to real experimental data.\"}]","Bridging Machine Learning and Sciences - Opportunities and Challenges | PDF",1785680600,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"bridging-machine-learning-and-sciences-opportunities-and-challenges","",{"@graph":35,"@context":84},[36,53,67],{"@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":52},"https://docshare.wps.com/document/bridging-machine-learning-and-sciences-opportunities-and-challenges/117974/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":22,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why is anomaly detection especially important for bridging machine learning and scientific applications?","Question",{"text":74,"@type":75},"Anomaly detection has long been studied, and deep neural networks have improved out-of-distribution detection for complex high-dimensional data. This capability is increasingly relevant in scientific settings.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What key differences separate general machine learning from tailored scientific applications?",{"text":79,"@type":75},"Differences span the nature of data, the inference process driven by experimental focus, benchmark limitations, uncertainty quantification needs, and challenges in generalization and robustness under simulation-to-real shifts.",{"name":81,"@type":72,"acceptedAnswer":82},"How does the document suggest improving model transfer from simulation to real scientific environments?",{"text":83,"@type":75},"It emphasizes adapting the workflow to the needs of the scientific domain, including accounting for dataset shift and precision requirements when transitioning from simulation data with labels to real experimental data.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]