{
	"event_id": "1139470",
	"eventinstance_id": "4197819",
	"calendar": {
		"id": 5902,
		"title": "TO-DO",
		"slug": "to-do",
		"url": "https://events.ucf.edu/calendar/5902/to-do/"
	},
	"id": "4197819",
	"title": "Mathematical Biology Seminar by Dr. Hsin\u002DHsiung \u0027Bill\u0027 Huang, UCF",
	"subtitle": null,
	"description": "\u003Cp\u003E\u003Cstrong\u003E\u003Ca href\u003D\u0022https://sciences.ucf.edu/mathbio/mathematical\u002Dbiology\u002Dseminar/\u0022 target\u003D\u0022_blank\u0022\u003EMathematical Biology Seminar\u003C/a\u003E\u003C/strong\u003E\u003C/p\u003E\u000A\u003Cp\u003E\u003Cspan\u003E\u003Ca href\u003D\u0022https://sciences.ucf.edu/sdmss/person/hsin\u002Dhsiung\u002Dhuang/\u0022 target\u003D\u0022_blank\u0022\u003E\u003Cstrong\u003EDr. Hsin\u002DHsiung \u0027Bill\u0027 Huang\u003C/strong\u003E\u003C/a\u003E (\u003C/span\u003E\u003Cspan\u003EUCF School of Data,\u0026nbsp\u003B\u003C/span\u003E\u003Cspan\u003EMathematical, and Statistical Sciences\u003C/span\u003E\u003Cspan\u003E)\u0026nbsp\u003B\u003C/span\u003E\u003Cspan\u003Ewill speak on \u0022\u003Cstrong\u003E\u003Cem\u003EBayesian Sparse Mixed\u002DResponse Species Distribution Models with Spatial Dependence\u003C/em\u003E\u003C/strong\u003E\u0022 at this week\u0027s seminar.\u003C/span\u003E\u003C/p\u003E\u000A\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C/strong\u003E \u003Cspan\u003EEcological monitoring programs increasingly record continuous, binary and count responses at common sites. We develop a Bayesian species distribution model for such mixed ecological responses, using a shared row\u002Dsparse coefficient matrix, P\u0026oacute\u003Blya\u0026ndash\u003BGamma augmentation for binary and negative\u002Dbinomial components, Gaussian updates for continuous components, and a low\u002Drank representation of spatial or spatio\u002Dtemporal structure. Raw environmental variables are expanded through a deterministic maximum\u002Dentropy\u002Dstyle feature dictionary before joint shrinkage is applied. The model preserves response\u002Dfamily\u002Dspecific likelihoods while allowing common environmental structure and residual dependence across responses. Controlled simulations show that the one\u002Dstep and screened two\u002Dstep estimators recover sparse coefficient rows and improve binary discrimination relative to response\u002Dwise sparse regressions when the data\u002Dgenerating mechanism contains shared sparsity and cross\u002Dresponse dependence. Heat maps, support\u002Drecovery summaries and residual maps diagnose sensitivity to spatial rank, count calibration and screening choices. The ecological analyses clarify the intended use. FishGlob motivates the full mixed\u002Dresponse survey problem, family\u002Dspecific diagnostics and dependence visualization. The African elephant occurrence\u002Dbackground benchmark shows why the Bernoulli special case must use the full spatial\u002Dsparse predictor rather than an environment\u002Donly sparse ablation: a strong spatial gradient cannot be represented adequately by sparse environmental features alone. After adding the spatial\u002Dbasis component and training\u002Dfold probability calibration, the proposed Bernoulli implementation has the strongest blocked cross\u002Dvalidation area under the receiver operating characteristic curve, Brier score and log score among the fitted elephant comparators. The method is most useful for high\u002Ddimensional mixed\u002Dresponse learning with interpretable shared sparsity and explicit spatial adjustment.\u003C/span\u003E\u003C/p\u003E\u000A\u003Cp\u003E\u003Cstrong\u003EShort Bio:\u003C/strong\u003E \u003Cspan\u003EDr. Hsin\u002DHsiung \u0026ldquo\u003BBill\u0026rdquo\u003B Huang is a Professor in Statistics and Data Science at the University of Central Florida (UCF). He earned his Ph.D. in Statistics from the University of Illinois at Chicago, an M.S. in Statistics from the Georgia Institute of Technology, an M.S. in Mathematics from National Taiwan University, and dual B.A./B.S. degrees (Economics and Mathematics) also from National Taiwan University. Dr. Huang\u0026rsquo\u003Bs research interests include Bayesian ultrahigh\u002Ddimensional variable selection, matrix\u002D and tensor\u002Dvariate regression, clustering, classification, and dimension reduction, with a focus on developing new statistical methods for complex, large\u002Dscale data. His recent projects encompass spatiotemporal modeling, novel neuroimaging reconstruction algorithms, and threat detection. His work has been supported by multiple grants from the National Science Foundation\u0026rsquo\u003Bs Algorithms for Threat Detection (ATD) program, where he serves as principal investigator (including NSF DMS\u002D1924792 and DMS\u002D2318925), as well as by the National Institutes of Health (NIH), where he is a co\u002Dinvestigator on a National Institute of Neurological Disorders and Stroke (NINDS) R01 award (1R01NS133094\u002D01A1).\u003C/span\u003E\u003C/p\u003E",
	"location": "MSB 318: Mathematical Sciences Building, Room 318",
	"location_url": "https://www.ucf.edu/location/mathematical\u002Dsciences\u002Dbuilding/",
	"virtual_url": null,
	"registration_link": null,
	"registration_info": null,
	"starts": "Tue, 01 Sep 2026 13:00:00 -0400",
	"ends": "Tue, 01 Sep 2026 14:00:00 -0400",
	"ongoing": "False",
	"category": "Speaker/Lecture/Seminar",
	"tags": ["UCF Statistics","Mathematical Biology Seminar","UCF Biology","UCF Mathematics"],
	"contact_name": "Erika Lin",
	"contact_phone": null,
	"contact_email": "erika.lin@ucf.edu",
	"url": "https://events.ucf.edu/event/4197819/mathematical-biology-seminar-by-dr-hsin-hsiung-bill-huang-ucf/"
}