{"id":2,"date":"2018-11-13T13:41:05","date_gmt":"2018-11-13T13:41:05","guid":{"rendered":"http:\/\/groups.cs.umass.edu\/infofusion\/?page_id=2"},"modified":"2024-03-28T04:41:08","modified_gmt":"2024-03-28T04:41:08","slug":"home","status":"publish","type":"page","link":"https:\/\/groups.cs.umass.edu\/infofusion\/","title":{"rendered":"Home"},"content":{"rendered":"<p><span style=\"font-weight: 400\">Welcome to the homepage of the Information Fusion Lab at the <a href=\"https:\/\/www.cics.umass.edu\/\">College of Information and Computer Sciences<\/a>, University of Massachusetts Amherst!<\/span><\/p>\n<p><span style=\"font-weight: 400\">We are a team of researchers at UMass Amherst CICS working on ML for multimodal data, including methods that support a wide range of biomedical applications. Our research includes a wide variety of topics including deep learning for the fusion of multi-resolution time series, images and structured information, the incorporation of domain knowledge or saliency in imaging and integration of multiple views for MRI analysis. Most recently, we introduced new methods for normalizing flows and transfer of causal models. Please see our research page for a full list of projects and our <\/span><a href=\"https:\/\/github.com\/Information-Fusion-Lab-Umass\"><span style=\"font-weight: 400\">GitHub page<\/span><\/a><span style=\"font-weight: 400\"> for code releases.<\/span><\/p>\n<p><span style=\"font-weight: 400\">We have an excellent team of talented graduate students and undergrads. If you are a UMass student and are interested in joining the group, or are a prospective external collaborator, please see <a href=\"https:\/\/groups.cs.umass.edu\/infofusion\/join\/\">this page<\/a>.<\/span><\/p>\n<p><span style=\"font-weight: 400\">Here is some recent news about our team:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400\">PhD student Deep Chakraborty was awarded the <a href=\"https:\/\/www.cics.umass.edu\/news\/latest-news\/edward-riseman-and-allen-hanson-scholarship-established\">Edward Riseman and Allen Hanson Scholarship<\/a><\/li>\n<li style=\"font-weight: 400\">PhD student Ke Xiao was awarded the <a href=\"https:\/\/umass.academicworks.com\/opportunities\/7074\">Paul Utgoff Memorial Scholarship<\/a><\/li>\n<li>The paper &#8216;<a href=\"https:\/\/invertibleworkshop.github.io\/accepted_papers\/pdfs\/40.pdf\">Normalizing Flows Across Dimensions<\/a>&#8216;, lead author Eddie Cunningham, was presented at the ICML 2020 INNF+ workshop<\/li>\n<li>The paper &#8216;<a href=\"https:\/\/biases-invariances-generalization.github.io\/pdf\/big_26.pdf\">Structure Mapping for Transferability of Causal Models<\/a>&#8216;, lead author Purva Pruthi, was presented at the ICML 2020 BIG workshop<\/li>\n<\/ul>\n<p>About the College of Information and Computer Sciences: <br \/>CICS is internationally recognized for its research activities and has one of the highest ranked and most competitive graduate programs in the nation. With over 40 faculty affiliated with the <a href=\"https:\/\/ds.cs.umass.edu\/\">Center for Data Science<\/a>, the College is distinguished by its culture of collaboration and leadership in multidisciplinary research. The department is #11 in AI and #20 in Computer Science in the US, according to the US News graduate schools ranking system. According to csrankings, CICS is in the <a href=\"http:\/\/csrankings.org\/#\/index?ai&amp;vision&amp;mlmining&amp;nlp&amp;ir\">top 10 universities in the US on AI<\/a> and <a href=\"http:\/\/csrankings.org\/#\/index?ai&amp;vision&amp;mlmining&amp;nlp&amp;ir&amp;world\">#16 in the world<\/a>.<\/p>\n<h2>Recent Publications<\/h2>\n<div class=\"teachpress_pub_list\"><form name=\"tppublistform\" method=\"get\"><a name=\"tppubs\" id=\"tppubs\"><\/a><\/form><div class=\"tablenav\"><div class=\"tablenav-pages\"><span class=\"displaying-num\">23 entries<\/span> <a class=\"page-numbers button disabled\">&laquo;<\/a> <a class=\"page-numbers button disabled\">&lsaquo;<\/a> 1 of 5 <a href=\"https:\/\/groups.cs.umass.edu\/infofusion\/?limit=2&amp;tgid=&amp;yr=&amp;type=&amp;usr=&amp;auth=&amp;tsr=#tppubs\" title=\"next page\" class=\"page-numbers button\">&rsaquo;<\/a> <a href=\"https:\/\/groups.cs.umass.edu\/infofusion\/?limit=5&amp;tgid=&amp;yr=&amp;type=&amp;usr=&amp;auth=&amp;tsr=#tppubs\" title=\"last page\" class=\"page-numbers button\">&raquo;<\/a> <\/div><\/div><table class=\"teachpress_publication_list\"><tr>\r\n                    <td>\r\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2025\">2025<\/h3>\r\n                    <\/td>\r\n                <\/tr><tr class=\"tp_publication tp_publication_conference\"><td class=\"tp_pub_info\"><p class=\"tp_pub_author\"> Shankar, Shiv;  Sinha, Ritwik;  Fiterau, Madalina<\/p><p class=\"tp_pub_title\"><a class=\"tp_title_link\" onclick=\"teachpress_pub_showhide('21','tp_links')\" style=\"cursor:pointer;\">Experimentation under Treatment Dependent Network Interference<\/a> <span class=\"tp_pub_type tp_  conference\">Conference<\/span> <\/p><p class=\"tp_pub_additional\"><span class=\"tp_pub_additional_booktitle\">41st Conference on Uncertainty in Artificial Intelligence, <\/span><span class=\"tp_pub_additional_volume\">vol. 286, <\/span><span class=\"tp_pub_additional_publisher\">PMLR, <\/span><span class=\"tp_pub_additional_year\">2025<\/span>.<\/p><p class=\"tp_pub_menu\"><span class=\"tp_resource_link\"><a id=\"tp_links_sh_21\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('21','tp_links')\" title=\"Show links and resources\" style=\"cursor:pointer;\">Links<\/a><\/span> | <span class=\"tp_bibtex_link\"><a id=\"tp_bibtex_sh_21\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('21','tp_bibtex')\" title=\"Show BibTeX entry\" style=\"cursor:pointer;\">BibTeX<\/a><\/span><\/p><div class=\"tp_bibtex\" id=\"tp_bibtex_21\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@conference{nokey,<br \/>\r\ntitle = {Experimentation under Treatment Dependent Network Interference},<br \/>\r\nauthor = {Shiv Shankar and Ritwik Sinha and Madalina Fiterau},<br \/>\r\neditor = {Silvia Chiappa},<br \/>\r\nurl = {https:\/\/proceedings.mlr.press\/v286\/shankar25a.html},<br \/>\r\nyear  = {2025},<br \/>\r\ndate = {2025-02-12},<br \/>\r\nurldate = {2025-02-12},<br \/>\r\nbooktitle = {41st Conference on Uncertainty in Artificial Intelligence},<br \/>\r\nvolume = {286},<br \/>\r\npages = {3787-3808},<br \/>\r\npublisher = {PMLR},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {conference}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('21','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_21\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"fas fa-globe\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/proceedings.mlr.press\/v286\/shankar25a.html\" title=\"https:\/\/proceedings.mlr.press\/v286\/shankar25a.html\" target=\"_blank\">https:\/\/proceedings.mlr.press\/v286\/shankar25a.html<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('21','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><tr>\r\n                    <td>\r\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2024\">2024<\/h3>\r\n                    <\/td>\r\n                <\/tr><tr class=\"tp_publication tp_publication_article\"><td class=\"tp_pub_info\"><p class=\"tp_pub_author\">Yunfei Lou Iman Deznabi, Abhinav Shaw<\/p><p class=\"tp_pub_title\"><a class=\"tp_title_link\" onclick=\"teachpress_pub_showhide('20','tp_links')\" style=\"cursor:pointer;\">Dynamic Clustering via Branched Deep Learning Enhances Personalization of Stress Prediction from Mobile Sensor Data<\/a> <span class=\"tp_pub_type tp_  article\">Journal Article<\/span> <\/p><p class=\"tp_pub_additional\"><span class=\"tp_pub_additional_in\">In: <\/span><span class=\"tp_pub_additional_journal\">Nature Scientific Reports, <\/span><span class=\"tp_pub_additional_volume\">vol. 14, <\/span><span class=\"tp_pub_additional_number\">no. 6631, <\/span><span class=\"tp_pub_additional_year\">2024<\/span>.<\/p><p class=\"tp_pub_menu\"><span class=\"tp_resource_link\"><a id=\"tp_links_sh_20\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('20','tp_links')\" title=\"Show links and resources\" style=\"cursor:pointer;\">Links<\/a><\/span> | <span class=\"tp_bibtex_link\"><a id=\"tp_bibtex_sh_20\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('20','tp_bibtex')\" title=\"Show BibTeX entry\" style=\"cursor:pointer;\">BibTeX<\/a><\/span><\/p><div class=\"tp_bibtex\" id=\"tp_bibtex_20\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@article{nokey,<br \/>\r\ntitle = {Dynamic Clustering via Branched Deep Learning Enhances Personalization of Stress Prediction from Mobile Sensor Data},<br \/>\r\nauthor = {Iman Deznabi, Yunfei Lou, Abhinav Shaw, Natcha Simsiri, Tauhidur Rahman, Madalina Fiterau},<br \/>\r\nurl = {https:\/\/www.nature.com\/articles\/s41598-024-56674-2},<br \/>\r\ndoi = {10.1038\/s41598-024-56674-2},<br \/>\r\nyear  = {2024},<br \/>\r\ndate = {2024-03-19},<br \/>\r\nurldate = {2024-03-19},<br \/>\r\njournal = {Nature Scientific Reports},<br \/>\r\nvolume = {14},<br \/>\r\nnumber = {6631},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {article}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('20','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_20\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"fas fa-globe\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/www.nature.com\/articles\/s41598-024-56674-2\" title=\"https:\/\/www.nature.com\/articles\/s41598-024-56674-2\" target=\"_blank\">https:\/\/www.nature.com\/articles\/s41598-024-56674-2<\/a><\/li><li><i class=\"ai ai-doi\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/dx.doi.org\/10.1038\/s41598-024-56674-2\" title=\"Follow DOI:10.1038\/s41598-024-56674-2\" target=\"_blank\">doi:10.1038\/s41598-024-56674-2<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('20','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><tr class=\"tp_publication tp_publication_workshop\"><td class=\"tp_pub_info\"><p class=\"tp_pub_author\">Peeyush Kumar Iman Deznabi, Madalina Fiterau<\/p><p class=\"tp_pub_title\"><a class=\"tp_title_link\" onclick=\"teachpress_pub_showhide('19','tp_links')\" style=\"cursor:pointer;\">Zero-shot micro-climate prediction with deep learning<\/a> <span class=\"tp_pub_type tp_  workshop\">Workshop<\/span> <\/p><p class=\"tp_pub_additional\"><span class=\"tp_pub_additional_year\">2024<\/span>.<\/p><p class=\"tp_pub_menu\"><span class=\"tp_abstract_link\"><a id=\"tp_abstract_sh_19\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('19','tp_abstract')\" title=\"Show abstract\" style=\"cursor:pointer;\">Abstract<\/a><\/span> | <span class=\"tp_resource_link\"><a id=\"tp_links_sh_19\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('19','tp_links')\" title=\"Show links and resources\" style=\"cursor:pointer;\">Links<\/a><\/span> | <span class=\"tp_bibtex_link\"><a id=\"tp_bibtex_sh_19\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('19','tp_bibtex')\" title=\"Show BibTeX entry\" style=\"cursor:pointer;\">BibTeX<\/a><\/span><\/p><div class=\"tp_bibtex\" id=\"tp_bibtex_19\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@workshop{Deznabi2024,<br \/>\r\ntitle = {Zero-shot micro-climate prediction with deep learning},<br \/>\r\nauthor = {Iman Deznabi, Peeyush Kumar, Madalina Fiterau},<br \/>\r\nurl = {https:\/\/arxiv.org\/abs\/2401.02665},<br \/>\r\nyear  = {2024},<br \/>\r\ndate = {2024-01-05},<br \/>\r\nurldate = {2024-01-05},<br \/>\r\njournal = {Tackling Climate Change with Machine Learning workshop NeurIPS (2023)},<br \/>\r\nabstract = {Weather station data is a valuable resource for climate prediction, however, its reliability can be limited in remote locations. To compound the issue, making local predictions often relies on sensor data that may not be accessible for a new, previously unmonitored location. In response to these challenges, we propose a novel zero-shot learning approach designed to forecast various climate measurements at new and unmonitored locations. Our method surpasses conventional weather forecasting techniques in predicting microclimate variables by leveraging knowledge extracted from other geographic locations.},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {workshop}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('19','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_abstract\" id=\"tp_abstract_19\" style=\"display:none;\"><div class=\"tp_abstract_entry\">Weather station data is a valuable resource for climate prediction, however, its reliability can be limited in remote locations. To compound the issue, making local predictions often relies on sensor data that may not be accessible for a new, previously unmonitored location. In response to these challenges, we propose a novel zero-shot learning approach designed to forecast various climate measurements at new and unmonitored locations. Our method surpasses conventional weather forecasting techniques in predicting microclimate variables by leveraging knowledge extracted from other geographic locations.<\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('19','tp_abstract')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_19\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"ai ai-arxiv\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/arxiv.org\/abs\/2401.02665\" title=\"https:\/\/arxiv.org\/abs\/2401.02665\" target=\"_blank\">https:\/\/arxiv.org\/abs\/2401.02665<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('19','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><tr>\r\n                    <td>\r\n                        <h3 class=\"tp_h3\" id=\"tp_h3_2023\">2023<\/h3>\r\n                    <\/td>\r\n                <\/tr><tr class=\"tp_publication tp_publication_workshop\"><td class=\"tp_pub_info\"><p class=\"tp_pub_author\"> Shankar, Shiv<\/p><p class=\"tp_pub_title\"><a class=\"tp_title_link\" onclick=\"teachpress_pub_showhide('22','tp_links')\" style=\"cursor:pointer;\">AI assisted Search for Atmospheric CO2 Capture<\/a> <span class=\"tp_pub_type tp_  workshop\">Workshop<\/span> <\/p><p class=\"tp_pub_additional\"><span class=\"tp_pub_additional_year\">2023<\/span>.<\/p><p class=\"tp_pub_menu\"><span class=\"tp_abstract_link\"><a id=\"tp_abstract_sh_22\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('22','tp_abstract')\" title=\"Show abstract\" style=\"cursor:pointer;\">Abstract<\/a><\/span> | <span class=\"tp_resource_link\"><a id=\"tp_links_sh_22\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('22','tp_links')\" title=\"Show links and resources\" style=\"cursor:pointer;\">Links<\/a><\/span> | <span class=\"tp_bibtex_link\"><a id=\"tp_bibtex_sh_22\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('22','tp_bibtex')\" title=\"Show BibTeX entry\" style=\"cursor:pointer;\">BibTeX<\/a><\/span><\/p><div class=\"tp_bibtex\" id=\"tp_bibtex_22\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@workshop{nokey,<br \/>\r\ntitle = {AI assisted Search for Atmospheric CO2 Capture},<br \/>\r\nauthor = {Shiv Shankar },<br \/>\r\nurl = {https:\/\/s3.us-east-1.amazonaws.com\/climate-change-ai\/papers\/neurips2023\/111\/paper.pdf},<br \/>\r\nyear  = {2023},<br \/>\r\ndate = {2023-12-01},<br \/>\r\nurldate = {2023-12-01},<br \/>\r\nabstract = {Carbon capture technologies is an important tool for mitigating climate change. In recent years, polymer membrane separation methods have emerged as a<br \/>\r\npromising technology for separating CO2 and other green house gases from the<br \/>\r\natmosphere. Designing new polymers for such tasks is quite difficult. In this work<br \/>\r\nwe look at machine learning based methods to search for new polymer designs<br \/>\r\noptimized for CO2 separation. An ensemble ML models is trained on a large<br \/>\r\ndatabase of molecules to predict permeabilities of CO2\/N2 and CO2\/O2 pairs. We<br \/>\r\nthen use search based optimization to discover new polymers that surpass existing<br \/>\r\npolymer designs. Simulations are then done to verify the predicted performance<br \/>\r\nof the new designs. Overall result suggests that ML based search can be used to<br \/>\r\ndiscover new polymers optimized for carbon capture},<br \/>\r\nhowpublished = {AI for Climate Change, Neurips 2023},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {workshop}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('22','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_abstract\" id=\"tp_abstract_22\" style=\"display:none;\"><div class=\"tp_abstract_entry\">Carbon capture technologies is an important tool for mitigating climate change. In recent years, polymer membrane separation methods have emerged as a<br \/>\r\npromising technology for separating CO2 and other green house gases from the<br \/>\r\natmosphere. Designing new polymers for such tasks is quite difficult. In this work<br \/>\r\nwe look at machine learning based methods to search for new polymer designs<br \/>\r\noptimized for CO2 separation. An ensemble ML models is trained on a large<br \/>\r\ndatabase of molecules to predict permeabilities of CO2\/N2 and CO2\/O2 pairs. We<br \/>\r\nthen use search based optimization to discover new polymers that surpass existing<br \/>\r\npolymer designs. Simulations are then done to verify the predicted performance<br \/>\r\nof the new designs. Overall result suggests that ML based search can be used to<br \/>\r\ndiscover new polymers optimized for carbon capture<\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('22','tp_abstract')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_22\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"fas fa-file-pdf\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/s3.us-east-1.amazonaws.com\/climate-change-ai\/papers\/neurips2023\/111\/paper.pdf\" title=\"https:\/\/s3.us-east-1.amazonaws.com\/climate-change-ai\/papers\/neurips2023\/111\/pape[...]\" target=\"_blank\">https:\/\/s3.us-east-1.amazonaws.com\/climate-change-ai\/papers\/neurips2023\/111\/pape[...]<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('22','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><tr class=\"tp_publication tp_publication_conference\"><td class=\"tp_pub_info\"><p class=\"tp_pub_author\"> Iman Deznabi, Madalina Fiterau<\/p><p class=\"tp_pub_title\"><a class=\"tp_title_link\" onclick=\"teachpress_pub_showhide('18','tp_links')\" style=\"cursor:pointer;\">MultiWave: Multiresolution Deep Architectures through Wavelet Decomposition for Multivariate Timeseries Forecasting and Prediction.<\/a> <span class=\"tp_pub_type tp_  conference\">Conference<\/span> <\/p><p class=\"tp_pub_additional\"><span class=\"tp_pub_additional_booktitle\">Conference on Health, Inference, and Learning (CHIL 2023), <\/span><span class=\"tp_pub_additional_volume\">vol. 209, <\/span><span class=\"tp_pub_additional_year\">2023<\/span>.<\/p><p class=\"tp_pub_menu\"><span class=\"tp_abstract_link\"><a id=\"tp_abstract_sh_18\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('18','tp_abstract')\" title=\"Show abstract\" style=\"cursor:pointer;\">Abstract<\/a><\/span> | <span class=\"tp_resource_link\"><a id=\"tp_links_sh_18\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('18','tp_links')\" title=\"Show links and resources\" style=\"cursor:pointer;\">Links<\/a><\/span> | <span class=\"tp_bibtex_link\"><a id=\"tp_bibtex_sh_18\" class=\"tp_show\" onclick=\"teachpress_pub_showhide('18','tp_bibtex')\" title=\"Show BibTeX entry\" style=\"cursor:pointer;\">BibTeX<\/a><\/span><\/p><div class=\"tp_bibtex\" id=\"tp_bibtex_18\" style=\"display:none;\"><div class=\"tp_bibtex_entry\"><pre>@conference{Deznabi2023,<br \/>\r\ntitle = {MultiWave: Multiresolution Deep Architectures through Wavelet Decomposition for Multivariate Timeseries Forecasting and Prediction.},<br \/>\r\nauthor = {Iman Deznabi, Madalina Fiterau},<br \/>\r\nurl = {https:\/\/proceedings.mlr.press\/v209\/deznabi23a\/deznabi23a.pdf},<br \/>\r\nyear  = {2023},<br \/>\r\ndate = {2023-06-25},<br \/>\r\nurldate = {2023-06-25},<br \/>\r\nbooktitle = {Conference on Health, Inference, and Learning (CHIL 2023)},<br \/>\r\nvolume = {209},<br \/>\r\nabstract = {The analysis of multivariate time series data is challenging due to the various frequencies of signal changes that can occur over both short and long terms. Furthermore, standard deep learning models are often unsuitable for such datasets, as signals are typically sampled at different rates. To address these issues, we introduce MultiWave, a novel framework that enhances deep learning time series models by incorporating components that operate at the intrinsic frequencies of signals. MultiWave uses wavelets to decompose each signal into subsignals of varying frequencies and groups them into frequency bands. Each frequency band is handled by a different component of our model. A gating mechanism combines the output of the components to produce sparse models that use only specific signals at specific frequencies. Our experiments demonstrate that MultiWave accurately identifies informative frequency bands and improves the performance of various deep learning models, including LSTM, Transformer, and CNN-based models, for a wide range of applications. It attains top performance in stress and affect detection from wearables. It also increases the AUC of the best-performing model by 5% for in-hospital COVID-19 mortality prediction from patient blood samples and for human activity recognition from accelerometer and gyroscope data. We show that MultiWave consistently identifies critical features and their frequency components, thus providing valuable insights into the applications studied.},<br \/>\r\nkeywords = {},<br \/>\r\npubstate = {published},<br \/>\r\ntppubtype = {conference}<br \/>\r\n}<br \/>\r\n<\/pre><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('18','tp_bibtex')\">Close<\/a><\/p><\/div><div class=\"tp_abstract\" id=\"tp_abstract_18\" style=\"display:none;\"><div class=\"tp_abstract_entry\">The analysis of multivariate time series data is challenging due to the various frequencies of signal changes that can occur over both short and long terms. Furthermore, standard deep learning models are often unsuitable for such datasets, as signals are typically sampled at different rates. To address these issues, we introduce MultiWave, a novel framework that enhances deep learning time series models by incorporating components that operate at the intrinsic frequencies of signals. MultiWave uses wavelets to decompose each signal into subsignals of varying frequencies and groups them into frequency bands. Each frequency band is handled by a different component of our model. A gating mechanism combines the output of the components to produce sparse models that use only specific signals at specific frequencies. Our experiments demonstrate that MultiWave accurately identifies informative frequency bands and improves the performance of various deep learning models, including LSTM, Transformer, and CNN-based models, for a wide range of applications. It attains top performance in stress and affect detection from wearables. It also increases the AUC of the best-performing model by 5% for in-hospital COVID-19 mortality prediction from patient blood samples and for human activity recognition from accelerometer and gyroscope data. We show that MultiWave consistently identifies critical features and their frequency components, thus providing valuable insights into the applications studied.<\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('18','tp_abstract')\">Close<\/a><\/p><\/div><div class=\"tp_links\" id=\"tp_links_18\" style=\"display:none;\"><div class=\"tp_links_entry\"><ul class=\"tp_pub_list\"><li><i class=\"fas fa-file-pdf\"><\/i><a class=\"tp_pub_list\" href=\"https:\/\/proceedings.mlr.press\/v209\/deznabi23a\/deznabi23a.pdf\" title=\"https:\/\/proceedings.mlr.press\/v209\/deznabi23a\/deznabi23a.pdf\" target=\"_blank\">https:\/\/proceedings.mlr.press\/v209\/deznabi23a\/deznabi23a.pdf<\/a><\/li><\/ul><\/div><p class=\"tp_close_menu\"><a class=\"tp_close\" onclick=\"teachpress_pub_showhide('18','tp_links')\">Close<\/a><\/p><\/div><\/td><\/tr><\/table><div class=\"tablenav\"><div class=\"tablenav-pages\"><span class=\"displaying-num\">23 entries<\/span> <a class=\"page-numbers button disabled\">&laquo;<\/a> <a class=\"page-numbers button disabled\">&lsaquo;<\/a> 1 of 5 <a href=\"https:\/\/groups.cs.umass.edu\/infofusion\/?limit=2&amp;tgid=&amp;yr=&amp;type=&amp;usr=&amp;auth=&amp;tsr=#tppubs\" title=\"next page\" class=\"page-numbers button\">&rsaquo;<\/a> <a href=\"https:\/\/groups.cs.umass.edu\/infofusion\/?limit=5&amp;tgid=&amp;yr=&amp;type=&amp;usr=&amp;auth=&amp;tsr=#tppubs\" title=\"last page\" class=\"page-numbers button\">&raquo;<\/a> <\/div><\/div><\/div>\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Welcome to the homepage of the Information Fusion Lab at the College of Information and Computer Sciences, University of Massachusetts Amherst! We are a team of researchers at UMass Amherst CICS working on ML for multimodal data, including methods that support a wide range of biomedical applications. Our research includes a wide variety of topics &hellip; <a href=\"https:\/\/groups.cs.umass.edu\/infofusion\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Home&#8221;<\/span><\/a><\/p>\n","protected":false},"author":11,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"open","template":"","meta":{"footnotes":""},"class_list":["post-2","page","type-page","status-publish","hentry","group-blog","no-sidebar","hfeed"],"_links":{"self":[{"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/pages\/2","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/users\/11"}],"replies":[{"embeddable":true,"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/comments?post=2"}],"version-history":[{"count":28,"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/pages\/2\/revisions"}],"predecessor-version":[{"id":306,"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/pages\/2\/revisions\/306"}],"wp:attachment":[{"href":"https:\/\/groups.cs.umass.edu\/infofusion\/wp-json\/wp\/v2\/media?parent=2"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}