{"680898":{"#nid":"680898","#data":{"type":"event","title":"CSIP Seminar | Adversarial evaluation of additive noise mechanisms via inferential minimum mean squared error (MMSE) estimation","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003EDate:\u003C\/strong\u003E\u0026nbsp;Monday, March 210, 2025\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ETime:\u003C\/strong\u003E\u0026nbsp;2:00 p.m.\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ELocation:\u0026nbsp;\u003C\/strong\u003ECentergy Building 5126\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ESpeaker:\u003C\/strong\u003E\u0026nbsp;Lalitha Sankar\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ESpeakers\u0027 Title:\u003C\/strong\u003E\u0026nbsp;Professor in the School of Electrical, Computer and Energy Engineering at Arizona State University\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ESeminar Title:\u003C\/strong\u003E\u0026nbsp;Adversarial evaluation of additive noise mechanisms via inferential minimum mean squared error (MMSE) estimation\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EAbstract:\u0026nbsp;\u003C\/strong\u003EWe propose a computationally feasible adversarial evaluation framework using minimum mean-squared error (MMSE) estimation and linear evaluation models. Our approach provides theoretical lower bounds on the true MMSE of inferring sensitive features from noisy observations of other correlated features. The bounds are in terms of the empirical MMSE under a restricted model\/hypothesis class and a decomposable error term capturing finite sample and approximation effects. For adversarial models using a sigmoid composed with linear hypothesis class, we derive closed-form bounds for classes of relationships between the private and non-private features, including linear mappings, binary symmetric channels, and class-conditional Gaussian models. Through empirical evaluation, we demonstrate that our linear model-based adversarial framework serves as a powerful yet tractable tool for MMSE-based adversarial evaluation that balances theoretical guarantees with practical efficiency.\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EBio: \u003C\/strong\u003ELalitha Sankar is a Professor in the School of Electrical, Computer and Energy Engineering at Arizona State University. She joined ASU as an assistant professor in fall of 2012, and was an associate professor from 2018-2023. She received a bachelor\u0027s degree from the Indian Institute of Technology, Bombay, a master\u0027s degree from the University of Maryland, and a doctorate from Rutgers University in 2007. Following her doctorate, Sankar was a recipient of a three-year Science and Technology Teaching Postdoctoral Fellowship from the Council on Science and Technology at Princeton University, following which she was an associate research scholar at Princeton. Prior to her doctoral studies, she was a senior member of technical staff at AT\u0026amp;T Shannon Laboratories. Sankar\u0027s research interests are at the intersection of information and data sciences including a background in signal processing, learning theory, and control theory with applications to the design of machine learning algorithms with algorithmic fairness, privacy, and robustness guarantees. Her research also applies such methods to complex networks including the electric power grid and healthcare systems.\u0026nbsp;\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003ELalitha Sankar, a Professor in the School of Electrical, Computer and Energy Engineering at Arizona State University, will present the Seminar, \u0022Adversarial evaluation of additive noise mechanisms via inferential minimum mean squared error estimation.\u0022\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"Lalitha Sankar, a Professor in the School of Electrical, Computer and Energy Engineering at Arizona State University, will present the Seminar, \u0022Adversarial evaluation of additive noise mechanisms via inferential minimum mean squared error estimation.\u0022"}],"uid":"36558","created_gmt":"2025-03-04 21:19:43","changed_gmt":"2025-03-04 21:25:04","author":"zwiniecki3","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2025-03-10T14:00:00-04:00","event_time_end":"2025-03-10T15:00:00-04:00","event_time_end_last":"2025-03-10T15:00:00-04:00","gmt_time_start":"2025-03-10 18:00:00","gmt_time_end":"2025-03-10 19:00:00","gmt_time_end_last":"2025-03-10 19:00:00","rrule":null,"timezone":"America\/New_York"},"location":" CSIP Library (Centergy Building, Room 5126)","extras":[],"groups":[{"id":"1255","name":"School of Electrical and Computer Engineering"}],"categories":[],"keywords":[{"id":"192224","name":"CSIP Seminar"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}