[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120244-en":3,"doc-seo-120244-105":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":20,"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},120244,7971461740909,"Levi","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Task relevant autoencoding enhances machine learning for human neuroscience","Human neuroscience research uses machine learning to extract low-dimensional neural representations linked to behavior, but common neuroimaging datasets are exemplar-poor while remaining high-dimensional, increasing the risk of overfitting in state-of-the-art models. TRACE (Task-Relevant Autoencoder via Classifier Enhancement) targets behaviorally relevant neural patterns rather than noise or unrelated factors. The method is benchmarked against standard autoencoders and other models on truncated datasets and evaluated on fMRI data from 59 subjects, showing improved classification accuracy and more interpretable task-relevant representations, indicating strong potential for behavioral neuroscience data analysis.","UC Irvine  \nUC Irvine Previously Published Works  \nTitle  \nTask relevant autoencoding enhances machine learning for human neuroscience.  \nPermalink  \n[https://escholarship.org/uc/item/2xh0x1ff](https://escholarship.org/uc/item/2xh0x1ff)  \nJournal  \nScientific Reports, 15(1)  \nAuthors  \nOrouji, Seyedmehdi Taschereau-Dumouchel, Vincent Cortese, Aurelio  \net al.  \nPublication Date  \n2025-01-08  \nDOI  \n10.1038/s41598-024-83867-6  \nPeer reviewed  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \n[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nTask relevant autoencoding enhances machine learning for human neuroscience  \nSeyedmehdi Orouji1􀀍, Vincent Taschereau-Dumouchel2,3, Aurelio Cortese4, Brian Odegaard5, Cody Cushing6, Mouslim Cherkaoui6, Mitsuo Kawato4, Hakwan Lau7 &  \nMeganA. K. Peters1,8􀀍  \nIn human neuroscience, machine learning can help reveal lower-dimensional neural representations relevant to subjects’ behavior. However, state-of-the-art models typically require large datasets to train, and so are prone to overfitting on human neuroimaging data that often possess few samples but many input dimensions. Here, we capitalized on the fact that the features we seek in human neuroscience are precisely those relevant to subjects’ behavior rather than noise or other irrelevant factors. We thus developed aTask-Relevant Autoencoder via Classifier Enhancement (TRACE) designed to identify behaviorally-relevant target neural patterns. We benchmarked TRACE against a standard autoencoder and other models for two severely truncated machine learning datasets (to match the data typically available in functional magnetic resonance imaging [fMRI] data for an individual subject), then evaluated all models on fMRI data from 59 subjects who observed animals and objects. TRACE outperformed alternative models nearly unilaterally, showing up to 12% increased classification accuracy and up to 56% improvement in discovering “cleaner”, task-relevant representations. These results showcase TRACE’s potential for a wide variety of data related to human behavior.  \nKeywords Human neuroscience, Machine learning, Dimensionality reduction, Task-relevant representation, fMRI, MVPA, Autoencoder  \nIn studying the human brain and human behavior, we often use machine learning methods to home in on the (ideally lower-dimensional1–4) representations contained in multivariate, feature-rich datasets. These data typically contain noisy, task-irrelevant signals5–7 that we would like to filter out using methods such as multivariate decoders8–11, various types of autoencoders, generative adversarial networks like InfoGAN12, or even principal components analysis (PCA)13–15. However, state-of-the-art machine learning methods typically require very large datasets to train while data for individual human subjects collected with methods such as functional magnetic resonance imaging (fMRI)5–7 are often severely limited in sample size16, 17 (i.e., have very few training exemplars compared to the dimensionality of the data) . Consequently, even our best, state-of-the-art methods are susceptible to overfitting on such neuroimaging data, reducing their predictive power and utility18–20. What’s more, parametric methods (such as PCA), which may better avoid the need for large training sets, by definition require specific assumptions regarding the nature of the dimensionality reduction process (e.g., the common assumption of linear dimensionality reduction) and thus are limited a priori to insights consistent with these parametric assumptions. One might hope to partially alleviate the data volume issue by functionally pooling data across participants using techniques such as hyperalignment21–24. However, such methods can introduce other challenges stemming from domain shift between individuals (i.e., statistical differences in voxels’ response distributions across subjects); such domain shift c","cbCaiuzfxQAwsrYm","https://ap.wps.com/l/cbCaiuzfxQAwsrYm","pdf",3900947,1,15,"English","en",105,"# Abstract\n# Background and Motivation\n## Data limitations in fMRI\n# Method\n## TRACE model and architecture\n# Evaluation\n## Benchmarks and datasets","[{\"question\":\"Why do existing machine learning models struggle with human neuroimaging data?\",\"answer\":\"They often require large training datasets, but fMRI data for individual subjects typically have very limited sample sizes relative to the number of input dimensions, which increases overfitting risk.\"},{\"question\":\"What is TRACE and what goal does it serve?\",\"answer\":\"TRACE is a task-relevant autoencoder designed to identify neural patterns that are relevant to subjects’ behavior rather than noise or other irrelevant factors.\"},{\"question\":\"How does TRACE perform compared with baseline models?\",\"answer\":\"On truncated machine-learning datasets and fMRI data from 59 subjects, TRACE outperformed alternative models nearly unilaterally, improving classification accuracy and enhancing discovery of cleaner, task-relevant representations.\"}]","Task relevant autoencoding enhances machine learning for human neuroscience | PDF",1785728960,38,{"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},"task-relevant-autoencoding-enhances-machine-learning-for-human-neuroscience","",{"@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/task-relevant-autoencoding-enhances-machine-learning-for-human-neuroscience/120244/",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","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why do existing machine learning models struggle with human neuroimaging data?","Question",{"text":75,"@type":76},"They often require large training datasets, but fMRI data for individual subjects typically have very limited sample sizes relative to the number of input dimensions, which increases overfitting risk.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is TRACE and what goal does it serve?",{"text":80,"@type":76},"TRACE is a task-relevant autoencoder designed to identify neural patterns that are relevant to subjects’ behavior rather than noise or other irrelevant factors.",{"name":82,"@type":73,"acceptedAnswer":83},"How does TRACE perform compared with baseline models?",{"text":84,"@type":76},"On truncated machine-learning datasets and fMRI data from 59 subjects, TRACE outperformed alternative models nearly unilaterally, improving classification accuracy and enhancing discovery of cleaner, task-relevant representations.","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"]