{"690738":{"#nid":"690738","#data":{"type":"event","title":"Ph.D. Defense Announcement - Yuichiro Aoyama","body":[{"value":"\u003Cdiv\u003E\u003Cstrong\u003ETitle: \u003C\/strong\u003EEntropic-Regularized Second-Order Dynamic Optimization\u003C\/div\u003E\u003Cdiv\u003E\u003Cstrong\u003EDate:\u003C\/strong\u003E\u0026nbsp;June 26, 2026\u003C\/div\u003E\u003Cdiv\u003E\u003Cstrong\u003ETime:\u003C\/strong\u003E\u0026nbsp;12:00 PM\u003C\/div\u003E\u003Cdiv\u003E\u003Cstrong\u003ELocation:\u003C\/strong\u003E\u0026nbsp;\u003Ca href=\u0022https:\/\/teams.microsoft.com\/meet\/22451334760459?p=xtoZ7nmnK1RtTcKEmR\u0022 id=\u0022OWA80802ac0-25e5-885c-374f-90469306498c\u0022 rel=\u0022noopener noreferrer\u0022 target=\u0022_blank\u0022 title=\u0022https:\/\/teams.microsoft.com\/meet\/22451334760459?p=xtoZ7nmnK1RtTcKEmR\u0022\u003ECoda Conference Room C1115\u003C\/a\u003E\u003C\/div\u003E\u003Cdiv\u003E\u0026nbsp;\u003C\/div\u003E\u003Cdiv\u003EYuichiro Aoyama\u003C\/div\u003E\u003Cdiv\u003EMachine Learning PhD Candidate\u003C\/div\u003E\u003Cdiv\u003ESchool of Aerospace Engineering\u003C\/div\u003E\u003Cdiv\u003EGeorgia Institute of Technology\u003C\/div\u003E\u003Cdiv\u003E\u0026nbsp;\u003C\/div\u003E\u003Cdiv\u003E\u003Cstrong\u003ECommittee\u003C\/strong\u003E\u003C\/div\u003E\u003Cdiv\u003EDr. Evangelos A. Theodorou (Advisor), School of Aerospace Engineering, Georgia Tech\u003C\/div\u003E\u003Cdiv\u003EDr. Glen Chou, School of Aerospace Engineering, Georgia Tech\u0026nbsp;\u003C\/div\u003E\u003Cdiv\u003EDr. Kyriakos G. Vamvoudakis, School of Aerospace Engineering, Georgia Tech\u003C\/div\u003E\u003Cdiv\u003EDr. Frank Dellaert, School of Interactive Computing, Georgia Tech\u003C\/div\u003E\u003Cdiv\u003EDr. Kenshiro Oguri, School of Aeronautics and Astronautics, Purdue University \u0026nbsp;\u003C\/div\u003E\u003Cdiv\u003E\u0026nbsp;\u003C\/div\u003E\u003Cdiv\u003E\u003Cstrong\u003EAbstract\u003C\/strong\u003E\u003C\/div\u003E\u003Cdiv\u003EThis dissertation addresses the challenge of performing optimal control for dynamical systems operating in non-convex cost landscapes arising from nonlinear dynamics and cluttered environments. While classical second-order methods like Differential Dynamic Programming (DDP) offer fast local convergence, they easily get trapped in local minima due to their reliance on local information. Although purely sampling-based methods bypass the locality issue, they suffer from high sample complexity and noisy decision-making. To bridge this gap, this work leverages the Maximum Entropy DDP (ME-DDP) framework, in which a structured exploration covariance naturally arises from entropic regularization, balancing second-order local exploitation with robust trajectory perturbation. We extend this mechanism beyond standard Shannon entropy to generalized representations, detailing the development of Tsallis entropy and Stein Variational DDP (SV-DDP) to maintain policy diversity without sacrificing optimization structure. Benchmarked against Model Predictive Path Integral (MPPI) variants within a Model Predictive Control (MPC) framework, the proposed algorithms demonstrate superior overall performance, with their practical robustness validated through hardware experiments on a quadrotor navigation task in cluttered environments.\u003C\/div\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EEntropic-Regularized Second-Order Dynamic Optimization\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"Entropic-Regularized Second-Order Dynamic Optimization"}],"uid":"36872","created_gmt":"2026-06-12 15:26:07","changed_gmt":"2026-06-12 15:33:34","author":"fkhan47","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2026-06-26T12:00:00-04:00","event_time_end":"2026-06-26T14:00:49-04:00","event_time_end_last":"2026-06-26T14:00:49-04:00","gmt_time_start":"2026-06-26 16:00:00","gmt_time_end":"2026-06-26 18:00:49","gmt_time_end_last":"2026-06-26 18:00:49","rrule":null,"timezone":"America\/New_York"},"location":"Coda Conference Room C1115 or Zoom Link","extras":[],"related_links":[{"url":"https:\/\/teams.microsoft.com\/meet\/22451334760459?p=xtoZ7nmnK1RtTcKEmR","title":""}],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"100811","name":"Phd Defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78771","name":"Public"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}