BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Penn Engineering Events - ECPv6.15.18//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Penn Engineering Events
X-ORIGINAL-URL:https://seasevents.nmsdev7.com
X-WR-CALDESC:Events for Penn Engineering Events
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20241103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20250309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20251102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20260308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20261101T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250903T153000
DTEND;TZID=America/New_York:20250903T163000
DTSTAMP:20260403T172656
CREATED:20250818T200954Z
LAST-MODIFIED:20250818T200954Z
UID:10008427-1756913400-1756917000@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Water–Hydrophobe Interfaces: From Debunking Myths to Boosting Global Food–Water–Climate Resilience" (Himanshu Mishra\, KAUST)
DESCRIPTION:Abstract: \nFirst\, I will outline our work on overcoming barriers to desert rehabilitation for urban greening\, landscaping\, and regional food security. In arid regions\, freshwater is scarce\, sandy soils lose water and fertilizers rapidly\, which stifles plant growth. Our team has developed two complementary solutions: (i) Superhydrophobic Sand (SHS)1 — a plastic-free\, bio-inspired mulch that cuts evaporative water loss from topsoil by ~80%; (ii) CarboSoil2 — an engineered biochar that greatly enhances nutrient and water retention in sandy soils. I will share our translational journey from materials invention and characterization to multi-year field trials with food crops and native plants in collaboration with industry partners. Results show SHS and CarboSoil can dramatically improve irrigation and fertilizer-use efficiency in hot\, dry\, sandy environments. Terraxy LLC\, the startup I co-founded\, is now scaling Carbosoil production to 6\,000 tons/year for regional sustainability projects (video link). \nNext\, I will discuss our investigation into controversial claims that hydrogen peroxide (H₂O₂) forms spontaneously at the air–water interface in microdroplets3-5 — a phenomenon that would upend fundamental physical chemistry and impact fields from cloud chemistry to green synthesis. We tested H₂O₂ formation in sprayed and condensed microdroplets under varied flow rates\, air compositions\, pH\, and salt content. Our findings show that the effect is not due to the air–water interface or microdroplet geometry6. Instead\, H₂O₂ originates from the reduction of dissolved oxygen at the solid–water interface7. Within a 50 nM detection limit\, no H₂O₂ was detected in oxygen-free conditions (more here). \n  \nReferences \n\nA. Gallo\, K. Odokonyero\, M. A. A. Mousa\, J. Reihmer\, S. Al-Mashharawi\, R. Marasco\, E. Manalastas\, M. J. L. Morton\, D. Daffonchio\, M. F. McCabe\, M. Tester and H. Mishra\, ACS Agricultural Science & Technology\, 2022\, 2\, 276-288.\nK. Odokonyero\, B. Vernooij\, B. Albar\, L. O. Exposito\, A. Alsamdani\, A. A. G. Haider\, N. V. H. Musskopf\, N. Kharbatia\, A. Gallo and H. Mishra\, Frontiers in Plant Science\, 2024\, Volume 15 – 2024.\nM. A. Mehrgardi\, M. Mofidfar and R. N. Zare\, Journal of the American Chemical Society\, 2022\, 144\, 7606-7609.\nJ. K. Lee\, K. L. Walker\, H. S. Han\, J. Kang\, F. B. Prinz\, R. M. Waymouth\, H. G. Nam and R. N. Zare\, P Natl Acad Sci USA\, 2019\, 116\, 19294-19298.\nR. A. LaCour\, J. P. Heindel\, R. Zhao and T. Head-Gordon\, J Am Chem Soc\, 2025\, 147\, 6299-6317.\nN. H. Musskopf\, A. Gallo\, P. Zhang\, J. Petry and H. Mishra\, The Journal of Physical Chemistry Letters\, 2021\, 12\, 11422-11429.\nM. A. Eatoo and H. Mishra\, Chem Sci\, 2024\, 15\, 3093-3103.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-water-hydrophobe-interfaces-from-debunking-myths-to-boosting-global-food-water-climate-resilience-himanshu-mishra-kaust/
LOCATION:Wu & Chen Auditorium
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250904T120000
DTEND;TZID=America/New_York:20250904T130000
DTSTAMP:20260403T172656
CREATED:20250825T231628Z
LAST-MODIFIED:20250825T231628Z
UID:10008456-1756987200-1756990800@seasevents.nmsdev7.com
SUMMARY:FOLDS Seminar: Positive random walks and positive-semidefinite random matrices
DESCRIPTION:On the real line\, a random walk that can only move in the positive direction is very unlikely to remain close to its starting point. After a fixed number of steps\, the left tail has a Gaussian profile\, under minimal assumptions. Remarkably\, the same phenomenon occurs when we consider a positive random walk on the cone of positive-semidefinite matrices. After a fixed number of steps\, the minimum eigenvalue is also described by a Gaussian model. \nThis talk introduces a new way to make this intuition rigorous. The methodology provides the solution to an open problem in computational mathematics about sparse random embeddings. The presentation is targeted at a general mathematical audience. \n  \nZoom link: https://upenn.zoom.us/j/98220304722
URL:https://seasevents.nmsdev7.com/event/folds-seminar-positive-random-walks-and-positive-semidefinite-random-matrices/
LOCATION:Amy Gutmann Hall\, Room 306\, 3317 Chestnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar,Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250904T140000
DTEND;TZID=America/New_York:20250904T150000
DTSTAMP:20260403T172656
CREATED:20250828T180248Z
LAST-MODIFIED:20250828T180248Z
UID:10008473-1756994400-1756998000@seasevents.nmsdev7.com
SUMMARY:BE Doctoral Dissertation Defense: Aoife O'Farrell
DESCRIPTION:The Department of Bioengineering at the University of Pennsylvania and Dr. Arjun Raj are pleased to announce the Doctoral Dissertation Defense of Aoife O’Farrell. \nTitle: Stimulus Specificity of Trained Immune Memory in Human Macrophages\nAdvisor: Dr. Arjun Raj\nDate and Time: Thursday\, September 4th at 2:00 PM\nLocation: Smilow 12-146AB\nZoom Link: https://upenn.zoom.us/j/94438324577 \nThe public is welcome to attend.
URL:https://seasevents.nmsdev7.com/event/be-doctoral-dissertation-defense-aoife-ofarrell/
CATEGORIES:Dissertation or Thesis Defense
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250905T110000
DTEND;TZID=America/New_York:20250905T120000
DTSTAMP:20260403T172656
CREATED:20250828T175551Z
LAST-MODIFIED:20250828T175551Z
UID:10008472-1757070000-1757073600@seasevents.nmsdev7.com
SUMMARY:BE Doctoral Dissertation Defense: Michael Yao
DESCRIPTION:The Department of Bioengineering at the University of Pennsylvania and Drs. Osbert Bastani and James Gee are pleased to announce the Doctoral Dissertation Defense of Michael Yao. \nTitle: Distributionally Robust Machine Intelligence\nAdvisors: Osbert Bastani\, PhD and James Gee\, PhD\nDate & Time: Friday\, September 5th at 11am\nLocation: Amy Gutmann Hall\, Room 414 (3317 Chestnut St\, Philadelphia\, PA 19104) \nRSVP Link: https://www.michaelsyao.com/defense \nZoom Link: https://pennmedicine.zoom.us/j/93390061157?pwd=gzaxVRmcZs0ASpUxBVpEtkMhMxTV2I.1 \nThe public is welcome to attend. \n 
URL:https://seasevents.nmsdev7.com/event/be-doctoral-dissertation-defense-michael-yao/
CATEGORIES:Dissertation or Thesis Defense
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250908T150000
DTEND;TZID=America/New_York:20250908T160000
DTSTAMP:20260403T172656
CREATED:20250717T155534Z
LAST-MODIFIED:20250717T155534Z
UID:10008412-1757343600-1757347200@seasevents.nmsdev7.com
SUMMARY:Future Proof Your Research With Rigor
DESCRIPTION:What’s the best way to avoid paper retractions and irreproducible results? Conduct rigorous research. Join us for a public talk with Ivan Oransky\, Co-Founder of Retraction Watch\, Anita Anita Bandrowski\, founder of The Research Resource Identification Initiative (RRID)\, and Jason Williams\, Assisant Director of the DNA Learning Center\, Cold Spring Harbor Laboratory\, on doing science that stands the test of time—and scrutiny. \nHouston Hall – Bodek Lounge\, 3417 Spruce Street Philadelphia\, PA 19104 \nRSVP here to save your spot! \n \n*This event is part of the C4R25 Conference – Sep 8-10 in Philadelphia – the world’s only conference dedicated to research rigor! Learn more at c4r.io/conference 
URL:https://seasevents.nmsdev7.com/event/future-proof-your-research-with-rigor/
LOCATION:Bodek Lounge\, Houston Hall\, 3417 Spruce St\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Faculty,Graduate,Student,Panel Discussion,Master's,Conference
ORGANIZER;CN="Community for Rigor":MAILTO:c4r@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250909T101500
DTEND;TZID=America/New_York:20250909T111500
DTSTAMP:20260403T172656
CREATED:20250903T142619Z
LAST-MODIFIED:20250903T142619Z
UID:10008496-1757412900-1757416500@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: MEAM Faculty Showcase
DESCRIPTION:Please join us on Tuesday\, September 9 for an overview of research being done in the MEAM Department\, hosted by MEAM Department Chair\, Dr. Kevin Turner. This is an excellent opportunity for current graduate students to learn about the breadth of work being done in MEAM. The following faculty will be presenting (not in order of presentation):
URL:https://seasevents.nmsdev7.com/event/meam-seminar-meam-faculty-showcase/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250910T120000
DTEND;TZID=America/New_York:20250910T131500
DTSTAMP:20260403T172656
CREATED:20250821T202705Z
LAST-MODIFIED:20250821T202705Z
UID:10008446-1757505600-1757510100@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "Rethinking Test-Time Thinking: From Token-Level Rewards to Robust Generative Agents"
DESCRIPTION:We present a unified perspective on test-time thinking as a lens for improving generative AI agents through finer-grained reward modeling\, data-centric reasoning\, and robust alignment. Beginning with GenARM\, we introduce an inductive bias for denser\, token-level reward modeling that guides generation during decoding\, enabling token-level alignment without retraining. While GenARM targets reward design\, ThinkLite-VL focuses on the data side of reasoning. It proposes a self-improvement framework that selects the most informative samples via MCTS-guided search\, yielding stronger visual reasoning with fewer labels. Taking this a step further\, MORSE-500 moves beyond selection to creation: it programmatically generates targeted\, controllable multimodal data to systematically probe and stress-test models’ reasoning abilities. We then interrogate a central assumption in inference-time alignment: Does Thinking More Always Help? Our findings reveal that increased reasoning steps can degrade performance–not due to better or worse reasoning per se\, but due to rising variance in outputs\, challenging the naive scaling paradigm. Finally\, AegisLLM applies test-time thinking in the service of security\, using an agentic\, multi-perspective framework to defend against jailbreaks\, prompt injections\, and unlearning attacks–all at inference time. Together\, these works chart a path toward generative agents that are not only more capable\, but more data-efficient\, introspective\, and robust in real-world deployment. \n  \nSeminar Recording: https://drive.google.com/file/d/13jOKuou0QzqkMo9QHEdoHA1nCIxOPsbm/view?usp=drive_link
URL:https://seasevents.nmsdev7.com/event/asset-seminar-rethinking-test-time-thinking-from-token-level-rewards-to-robust-generative-agents/
LOCATION:Amy Gutmann Hall\, Room 414\, 3333 Chestnut Street\, Philadelphia\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="AI-enabled Systems%3A Safe%2C Explainable%2C and Trustworthy (ASSET) Center":MAILTO:asset-info@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250910T150000
DTEND;TZID=America/New_York:20250910T160000
DTSTAMP:20260403T172656
CREATED:20250902T163122Z
LAST-MODIFIED:20250902T163122Z
UID:10008489-1757516400-1757520000@seasevents.nmsdev7.com
SUMMARY:Fall 2025 GRASP SFI: Tairan He\, Carnegie Mellon University & NVIDIA\, “Scalable Sim-to-Real Learning for General-Purpose Humanoid Skills”
DESCRIPTION:This is a hybrid event with in-person attendance in Levine 307 and virtual attendance via Zoom.  \nABSTRACT\nHumanoids represent the most versatile robotic platform\, capable of walking\, manipulating\, and collaborating with people in human-centered environments. Yet\, despite recent advances\, building humanoids that can operate reliably in the real world remains a fundamental challenge. Progress has been hindered by difficulties in whole-body control\, robust perceptive reasoning\, and bridging the sim-to-real gap.\n\nIn this talk\, I will discuss how scalable simulation and learning can systematically overcome these barriers. I will present a research trajectory that advances humanoid capabilities along three dimensions:\n\n1. Sim-to-Real Control: From real-time teleoperation (H2O) to dexterous loco-manipulation (OmniH2O)\, to a versatile generalist controller (HOVER)\, and agile transfer via dynamics alignment (ASAP)\, these works demonstrate increasingly dexterous and adaptable control.\n2. Sim-to-Real Perception: With ABS\, we show that robust real-world locomotion requires tightly coupling exteroceptive and proprioceptive sensing—shifting sim-to-real learning from blind skill execution to perception-driven control.\n3. Future Directions: I will outline potential next steps in (1) Perceptive Loco-Manipulation: end-to-end visuomotor policies unifying perception\, locomotion\, and manipulation—and (2) Real-to-Sim Evaluation: using high-fidelity simulators environments to provide consistent evaluation protocols and benchmarks for real-world policy evaluation.\n\nTaken together\, these works explore how far scalable sim-to-real learning can advance humanoid capabilities across control and perception. While sim-to-real is not the only path toward reliable humanoids\, this talk aims to test its limits—probing how scaling simulation and learning can push humanoids closer to functioning as capable partners in real-world environments.
URL:https://seasevents.nmsdev7.com/event/fall-2025-grasp-sfi-tairan-he/
LOCATION:Levine 307\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="General Robotics%2C Automation%2C Sensing and Perception (GRASP) Lab":MAILTO:grasplab@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250910T153000
DTEND;TZID=America/New_York:20250910T163000
DTSTAMP:20260403T172656
CREATED:20250818T201347Z
LAST-MODIFIED:20250818T201347Z
UID:10008428-1757518200-1757521800@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Non-equilibrium Dynamics of Lipid Vesicles using Automated Flow Control" (Charles Schroeder\, Princeton University)
DESCRIPTION:Abstract: \nVesicles are membrane-bound compartments that play a central role in biology. Despite recent progress\, the dynamics of single- and multi-component lipid vesicles are not fully understood\, particularly far from equilibrium where complex nonspherical shapes undergo large deformations in flow. In this talk\, I will present recent work from our group on the non-equilibrium dynamics of lipid vesicles in precisely defined flows using the Stokes trap – a new method that enables full 3D control of position and orientation of molecules or particles using active feedback control\, without the need for external optical\, magnetic\, or electric fields. After characterizing equilibrium properties including bending modulus and membrane tension\, we study vesicle deformation as a function of dimensionless flow strength (capillary number\, Ca) and vesicle deflation (reduced volume). Our results show that vesicles are remarkably deformable\, exhibiting reversible shape changes with aspect ratios exceeding 20 in repeated stretch-relax cycles in the bending-dominated regime. Single-component vesicles show a rich variety of shapes and conformations\, including asymmetric and symmetric dumbbells\, in addition to pearling\, wrinkling\, and buckling instabilities\, depending on membrane properties. Based on these observations\, we construct a detailed flow-phase diagram for vesicles in extensional flow\, and we further analyze transient stretching and relaxation dynamics. Two distinct relaxation processes emerge for deformed vesicles\, including a fast relaxation process corresponding to bending modes and a slow process governed by membrane tension relaxation. Finally\, we study vesicle shape dynamics in time-dependent large-amplitude oscillatory flows\, revealing three distinct dynamical regimes – pulsating\, reorienting\, and symmetrical deformations – arising from the competition between flow and membrane deformation timescales. Together\, these results provide new insight into flow-driven shape instabilities for lipid vesicles using new methods in flow automation.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-non-equilibrium-dynamics-of-lipid-vesicles-using-automated-flow-control-charles-schroeder-princeton-university/
LOCATION:Wu & Chen Auditorium
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250911T120000
DTEND;TZID=America/New_York:20250911T130000
DTSTAMP:20260403T172656
CREATED:20250828T163935Z
LAST-MODIFIED:20250828T163935Z
UID:10008468-1757592000-1757595600@seasevents.nmsdev7.com
SUMMARY:FOLDS seminar: Algorithmic stability for regression and classification
DESCRIPTION:In a supervised learning setting\, a model fitting algorithm is unstable if small perturbations to the input (the training data) can often lead to large perturbations in the output (say\, predictions returned by the fitted model). Algorithmic stability is a desirable property with many important implications such as generalization and robustness\, but testing the stability property empirically is known to be impossible in the setting of complex black-box models. In this work\, we establish that bagging any black-box regression algorithm automatically ensures that stability holds\, with no assumptions on the algorithm or the data. Furthermore\, we construct a new framework for defining stability in the context of classification\, and show that using bagging to estimate our uncertainty about the output label will again allow stability guarantees for any black-box model. This work is joint with Jake Soloff and Rebecca Willett. \n  \nZoom link: https://upenn.zoom.us/j/98220304722
URL:https://seasevents.nmsdev7.com/event/folds-seminar-algorithmic-stability-for-regression-and-classification/
LOCATION:Amy Gutmann Hall\, Room 414\, 3333 Chestnut Street\, Philadelphia\, 19104\, United States
CATEGORIES:Seminar,Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250911T150000
DTEND;TZID=America/New_York:20250911T170000
DTSTAMP:20260403T172656
CREATED:20250820T171923Z
LAST-MODIFIED:20250820T171923Z
UID:10008440-1757602800-1757610000@seasevents.nmsdev7.com
SUMMARY:ESE PhD Thesis Defense - "Design and Characterization of AlScN-Based Ferroelectric Devices for Non-Volatile Memory Applications"
DESCRIPTION:Ferroelectric materials have emerged as promising candidates for next-generation non-volatile memory technologies due to their intrinsic remnant polarization and fast switching dynamics. Among them\, aluminum scandium nitride (AlScN) stands out for its CMOS compatibility\, low growth temperature\, exceptional thermal stability and high remnant polarization compared to conventional perovskite and hafnia-based ferroelectrics. \nIn this dissertation\, I present a comprehensive study of AlScN-based ferroelectric devices for non-volatile memory applications. I first demonstrate ferroelectric field-effect transistors (FeFETs) by integrating AlScN with 2D materials\, where the Sc composition in AlScN is controlled. Through systematic electrical characterization\, including current–voltage (I–V)\, polarization–voltage (P–V) loops\, and pulsed positive up negative down (PUND) measurements\, I elucidate device physics\, endurance\, and wake-up/fatigue behavior. By engineering the contact interface\, I further realize n-type\, p-type\, and ambipolar FeFETs with enhanced ferroelectric gating efficiency and substantially improved on-state current. \nNext\, I investigate highly scaled AlScN-based ferroelectric diodes (Fe-diodes) and highlight their potential for high-density memory integration. Taken together\, these studies establish AlScN as a robust material platform for ferroelectric device technologies\, bridging fundamental materials science with device- and circuit-level opportunities in memory and beyond.
URL:https://seasevents.nmsdev7.com/event/ese-phd-thesis-defense-design-and-characterization-of-alscn-based-ferroelectric-devices-for-non-volatile-memory-applications/
LOCATION:Room 221\, Singh Center for Nanotechnology\, 3205 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Dissertation or Thesis Defense
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250912T120000
DTEND;TZID=America/New_York:20250912T133000
DTSTAMP:20260403T172656
CREATED:20250903T165703Z
LAST-MODIFIED:20250903T165703Z
UID:10008497-1757678400-1757683800@seasevents.nmsdev7.com
SUMMARY:ESE PhD Thesis Defense - "Learning-based Safe and Robust Control for Multi-Agent Systems"
DESCRIPTION:AI-enabled systems have become ubiquitous and integral to safety-critical domains\, e.g.\, autonomous vehicles and aerial robotics. Despite promising empirical results\, decision-making processes for critical systems incorporating AI components require careful consideration\, as failures may have catastrophic consequences. One key challenge is that various uncertainties will inevitably arise from system limitations\, black-box models\, or environmental factors\, and inaccurate estimation of intrinsic uncertainties or failure to account for other agents in the environment can lead to hazardous behaviors. \nIn this dissertation\, we study how to develop safe and robust learning-based control policies under various uncertainties. In particular\, it explores how tools from statistics\, game theory and formal methods can empower uncertainty quantification\, adaptation to other agents\, and robust policy synthesis. The first part focuses on safe learning and control multi-agent systems\, where we show how to develop safe\, robust\, and adaptive control strategies in safety-critical systems when encountering other agents. The second part studies how to synthesize safe perception-based control policy for robotic systems under uncertainties.
URL:https://seasevents.nmsdev7.com/event/ese-phd-thesis-defense-learning-based-safe-and-robust-control-for-multi-agent-systems/
LOCATION:WYSIWYG
CATEGORIES:Dissertation or Thesis Defense
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250912T130000
DTEND;TZID=America/New_York:20250912T140000
DTSTAMP:20260403T172656
CREATED:20250828T151507Z
LAST-MODIFIED:20250828T151507Z
UID:10008467-1757682000-1757685600@seasevents.nmsdev7.com
SUMMARY:ESE PhD Thesis Defense - "Magnetostatic Surface Wave and Surface Acoustic Wave Devices for Tunable and Energy Efficient Radio Frequency Filters"
DESCRIPTION:Tunable and energy-efficient filters are key components in modern wireless communication\, where RF front-end systems must operate across multiple frequency bands while minimizing power consumption. \nThis thesis focuses on the design and fabrication of miniature\, narrowband\, tunable bandpass and bandstop filters based on magnetostatic waves (MSW) in yttrium iron garnet (YIG) waveguides. A zero-static-power magnetic bias circuit is used to tune the filter’s center frequency\, but the compact size of the magnetic bias circuit imposes stringent limits on the YIG waveguide dimensions. To address this challenge\, microfabricated YIG thin films with aluminum meander-line transducers were developed. These designs improve the resonator figure of merit\, reduce insertion loss\, and enhance coupling. Increasing the YIG thickness further improves skirt selectivity\, lowers propagation loss\, and increases power handling. By integrating wideband tunability with nonreciprocal behavior\, a single device can replace multiple RF switches\, filters\, and isolators\, simultaneously controlling passband selection and directional isolation. This integration simplifies RF front-end design by reducing the number of required components. \nIn parallel\, this thesis introduces a high-frequency surface acoustic wave (SAW) platform using aluminum scandium nitride (AlScN) on 4H-silicon carbide (SiC). This material system combines high sound velocity\, high thermal conductivity\, and strong piezoelectric response. Furthermore\, the acoustoelectric effect (AE) was harnessed to achieve nonreciprocal RF amplification. A proof-of-concept AE delay line was realized by integrating Sezawa-mode SAWs in AlScN/SiC with ion-implanted SiC\, supporting future applications for energy efficient RF amplifiers.
URL:https://seasevents.nmsdev7.com/event/ese-phd-thesis-defense-magnetostatic-surface-wave-and-surface-acoustic-wave-devices-for-tunable-and-energy-efficient-radio-frequency-filters/
LOCATION:Greenberg Lounge (Room 114)\, Skirkanich Hall\, 210 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Dissertation or Thesis Defense
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250913T100000
DTEND;TZID=America/New_York:20250913T110000
DTSTAMP:20260403T172656
CREATED:20250828T174856Z
LAST-MODIFIED:20250828T174856Z
UID:10008471-1757757600-1757761200@seasevents.nmsdev7.com
SUMMARY:BE Doctoral Dissertation Defense: Brian-Tinh D. Vu
DESCRIPTION:The Department of Bioengineering at the University of Pennsylvania and Dr. Chamith Rajapakse are pleased to announce the Doctoral Dissertation Defense of Brian-Tinh D. Vu. \nTitle: “Rapid\, High-resolution\, and Signal-efficient Methods for the Clinical Translation of Bone Magnetic Resonance Imaging”\nDate: Wednesday\, September 17th\nTime: 10:00 AM\nLocation: BRB 1412 \nZoom: https://upenn.zoom.us/j/99746435550?pwd=PSZPcdjQznAxQK4TRsyydCi0IQE52L.1 \nThe public is welcome to attend.
URL:https://seasevents.nmsdev7.com/event/be-doctoral-dissertation-defense-brian-tinh-d-vu/
CATEGORIES:Dissertation or Thesis Defense
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250915T123000
DTEND;TZID=America/New_York:20250915T143000
DTSTAMP:20260403T172656
CREATED:20250904T151234Z
LAST-MODIFIED:20250904T151234Z
UID:10008498-1757939400-1757946600@seasevents.nmsdev7.com
SUMMARY:BE Doctoral Dissertation Defense: Georgios Mentzelopoulos
DESCRIPTION:The Department of Bioengineering at the University of Pennsylvania and Dr. Flavia Vitale are pleased to announce the Doctoral Dissertation Defense of Georgios Mentzelopoulos. \nTitle: From thoughts to actions: cracking the neural code across scales and modalities\nAdvisor: Dr. Flavia Vitale\nDate: Monday\, September 15th\nTime: 12:30 PM- 2:30 PM\nLocation: 225 Towne Building \nThe public is welcome to attend.
URL:https://seasevents.nmsdev7.com/event/be-doctoral-dissertation-defense-georgios-mentzelopoulos/
CATEGORIES:Dissertation or Thesis Defense
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250916T101500
DTEND;TZID=America/New_York:20250916T111500
DTSTAMP:20260403T172656
CREATED:20250819T175647Z
LAST-MODIFIED:20250819T175647Z
UID:10008436-1758017700-1758021300@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Micro-surgical Tools for Dissecting Cells and Tissues"
DESCRIPTION:Wound healing is an essential biological process for maintaining homeostasis and\, ultimately\, for survival. We investigate the mechanisms underlying extreme wound healing in Stentor coeruleus\, a single-celled organism\, capable of recovering from drastic membrane wounds exceeding half of the cell surface. This talk focuses on our recent effort on developing a microfluidic platform for the manipulation and reproducible wounding of the cell. We demonstrate a microfluidic “guillotine” for bisecting cells in a continuous flow\, and a “SMORES” platform to immobilize the cells for laser ablation. We further discuss the extension of the tool for dissecting live tumors to generate organoids that preserve the tumor immune microenvironment for applications including the testing of immunotherapy on a chip.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-micro-surgical-tools-for-dissecting-cells-and-tissues/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250917T120000
DTEND;TZID=America/New_York:20250917T131500
DTSTAMP:20260403T172656
CREATED:20250821T202941Z
LAST-MODIFIED:20250821T202941Z
UID:10008448-1758110400-1758114900@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "Symbolic Reasoning in the Age of Large Language Models"
DESCRIPTION:Today\, reasoning is commonly interpreted as large language models generating chains of thought. Yet historically\, AI reasoning had a very different meaning: executing algorithms that manipulated symbols to perform logical or probabilistic deduction and derive definite answers to questions about knowledge. In this talk\, I show that such old-fashioned ideas are very relevant to reasoning with large language models today. In particular\, I will demonstrate that integrating symbolic reasoning algorithms directly into the architecture of language models enables state-of-the-art capabilities in controllable text generation\, alignment\, and mathematical reasoning. These capabilities are built on top of tractable probabilistic circuit models that approximate the distribution of the large language model’s future behavior\, and allow for efficient reasoning on the GPU. I will further show that the same ideas naturally extend to neurosymbolic offline reinforcement learning and image diffusion. \n  \nSeminar Recording: https://drive.google.com/file/d/1PF5GxivBHKiloFdKTv59NMae5dtXDzSk/view?usp=sharing \n 
URL:https://seasevents.nmsdev7.com/event/asset-seminar-title-tbd-2/
LOCATION:Amy Gutmann Hall\, Room 414\, 3333 Chestnut Street\, Philadelphia\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="AI-enabled Systems%3A Safe%2C Explainable%2C and Trustworthy (ASSET) Center":MAILTO:asset-info@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250917T153000
DTEND;TZID=America/New_York:20250917T163000
DTSTAMP:20260403T172656
CREATED:20250818T201637Z
LAST-MODIFIED:20250818T201637Z
UID:10008429-1758123000-1758126600@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Natural Product Research for Human Health and Biocatalysis" (Wenjun Zhang\, UC Berkeley)
DESCRIPTION:Abstract: \nNatural products are historically great sources of human medicines\, but their impacts are well beyond their use as drugs. In addition to well-known producers like environmental microbes and plants\, the human microbiome is an emerging source of new natural products that often correlate with health or disease. There is an urgent need for the experimental characterization of these abundant\, yet poorly understood\, molecules and the downstream socio-chemical relationships they mediate. In the first half of my lecture\, I will introduce efforts from my group on the discovery and functional study of new bioactive natural products\, particularly from the human oral microbiome. In the second half\, I will discuss our work on understanding natural product biosynthesis\, which makes it possible to produce more bioactive natural products and their analogs via synthetic biology and yield useful enzymes for biocatalysis. Our recent efforts on discovering and characterizing N-N bond-forming enzymes will be highlighted.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-natural-product-research-for-human-health-and-biocatalysis-wenjun-zhang-uc-berkeley/
LOCATION:Wu & Chen Auditorium
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250918T103000
DTEND;TZID=America/New_York:20250918T120000
DTSTAMP:20260403T172656
CREATED:20250810T210321Z
LAST-MODIFIED:20250810T210321Z
UID:10008423-1758191400-1758196800@seasevents.nmsdev7.com
SUMMARY:MSE Seminar: 1D Topological Systems for Next-Generation Electronics - Judy J. Cha - Cornell University
DESCRIPTION:Topological nanowires\, topological materials confined in one dimension (1D)\, hold great promise for robust and scalable quantum computing and low-dissipation interconnect applications\, which will transform current computing technologies. To do so\, research in topological nanowires must continue to improve their synthesis and properties. \nIn this talk\, I will discuss my group’s efforts to develop a high throughput and precision synthesis method to fabricate 1D topological systems. We employ recently developed thermomechanical nanomolding to extrude single crystal nanowires of topological materials with controlled diameter. I will highlight our transport studies on topological semimetal nanowires for their potential application as extremely scaled\, low-resistance interconnects. We demonstrate that the resistivity scaling of topological semimetal nanowires is superior to those of the state-of-the-art Cu interconnects and Cu alternative metals\, presenting them as viable candidates for the low-resistance interconnect applications.
URL:https://seasevents.nmsdev7.com/event/mse-seminar-1d-topological-systems-for-next-generation-electronics-judy-j-cha-cornell-university/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Materials Science and Engineering":MAILTO:johnruss@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250918T120000
DTEND;TZID=America/New_York:20250918T130000
DTSTAMP:20260403T172656
CREATED:20250828T165113Z
LAST-MODIFIED:20250828T165113Z
UID:10008469-1758196800-1758200400@seasevents.nmsdev7.com
SUMMARY:FOLDS seminar: Heaviside Composite Optimization: A new paradigm of optimization
DESCRIPTION:Zoom link: https://upenn.zoom.us/j/98220304722 \n  \nA Heaviside function is an indicator function of a semi-infinite interval. A Heaviside composite function is a Heaviside function composed with a multivariate function that may be nonsnooth. This talk presents a touch of this novel class of discontinuous optimization problems that borders on continuous and discrete optimization. Among the rich applications of such problems\, we mention one pertaining to classification and treatment with constraints solvable by an elementary yet novel progressive integer programming (PIP) method that can be applied broadly to many other problems\,​ including indefinite quadratic programs and linear programs with linear​ complementarity constraints.  Computational results demonstrate the excellent​ performance of the PIP idea and its superiority over a full integer programming approach.
URL:https://seasevents.nmsdev7.com/event/folds-seminar-tba/
LOCATION:Amy Gutmann Hall\, Room 414\, 3333 Chestnut Street\, Philadelphia\, 19104\, United States
CATEGORIES:Seminar,Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250919T080000
DTEND;TZID=America/New_York:20250919T170000
DTSTAMP:20260403T172656
CREATED:20250807T164240Z
LAST-MODIFIED:20250807T164240Z
UID:10008422-1758268800-1758301200@seasevents.nmsdev7.com
SUMMARY:AI Research Mixer 2025
DESCRIPTION:Event Date: Friday\, September 19\, 2025 \nTime: 8 a.m. to 5 p.m. \nEvent Location: Amy Gutmann Hall Auditorium \n\nHosted by: \nASSET Center for Trustworthy AI \nIDEAS Innovation in Data Engineering and Science \nThe Warren Center for Network & Data Sciences \nDDDI Data Driven Discovery Initiative \nPenn AI \n\n\n\n\n\n\nEvent Description: \nPlease join us for a full day of faculty talks\, poster presentations\, and networking opportunities with members of ASSET\, IDEAS\, Warren\, DDDI and Penn AI to learn about the wide range of AI research happening here at Penn. \nAll Penn community members who have interest in AI are welcome! \n\nCatering will be provided throughout the day\, including a light reception following the final talks. \n\n\n\n\nClick here for the event schedule. \n\n 
URL:https://seasevents.nmsdev7.com/event/ai-research-mixer-2025/
LOCATION:Amy Gutmann Hall\, Auditorium\, 3333 Chestnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Conference,Symposium
ORGANIZER;CN="The Warren Center":MAILTO:Lhoot@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250919T103000
DTEND;TZID=America/New_York:20250919T114500
DTSTAMP:20260403T172656
CREATED:20250902T201243Z
LAST-MODIFIED:20250902T201243Z
UID:10008492-1758277800-1758282300@seasevents.nmsdev7.com
SUMMARY:[VIRTUAL SPEAKER]: Fall 2025 GRASP on Robotics: Max Welling\, University of Amsterdam & CuspAI\, "On the Role of Uncertainty in Automating the Scientific  Process with AI"
DESCRIPTION:This event will have a VIRTUAL SPEAKER. Streaming will be in-person only in Wu and Chen.  \nABSTRACT\nWe are in the middle of a transformation of the scientific process. AI is now rapidly automating scientific discovery by accelerating scientific simulation and optimisation. For example\, we can now shortcut expensive quantum mechanical calculations such as DFT with much faster machine learned interatomic potentials and forces\, numerically “solve” PDEs with neural networks\, or accelerate experimental design with Bayesian optimisation. However\, in both applications of AI the role of uncertainty is key because ML models can only be trusted in regions of input space where it has seen data. Extrapolating too far away from the data runs the risk of a large prediction error that goes unnoticed. In this talk I will first provide a high level overview of the increasingly important role of AI in science\, and then proceed with introducing a new efficient Variational Bayesian method for uncertainty quantification. I will show the effectiveness of this method on PDE solving and molecular generation (BARNN) and on ML interatomic potentials and forces (BLIP).
URL:https://seasevents.nmsdev7.com/event/virtual-speaker-fall-2025-grasp-on-robotics-max-welling-university-of-amsterdam-cuspai-on-the-role-of-uncertainty-in-automating-the-scientific-process-with-ai/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="General Robotics%2C Automation%2C Sensing and Perception (GRASP) Lab":MAILTO:grasplab@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250919T140000
DTEND;TZID=America/New_York:20250919T150000
DTSTAMP:20260403T172656
CREATED:20250829T143727Z
LAST-MODIFIED:20250829T143727Z
UID:10008481-1758290400-1758294000@seasevents.nmsdev7.com
SUMMARY:PICS Colloquium: Enabling computationally efficient first-principles kinetic simulations in nanoporous catalysts using machine learning and data science with Brandon Bukowski
DESCRIPTION:Machine learning tools have tremendous potential to accelerate computationally complex physics-based simulations. One example is the need to accelerate materials discovery through first-principles Density Functional Theory (DFT) calculations. \nThis seminar will encompass how machine learning interatomic potentials accelerate the discovery of crystalline nanoporous solids such as zeolites and metal-organic frameworks (MOFs) that are employed in a wide range of catalytic processes due in part to their tunable micro-environments. Kinetics at intracrystalline sites can be modified by changing pore size\, pore architecture\, or polarity. These environments impart shape-selectivity that preferentially stabilizes transition states\, but the large design space including pore architecture\, polarity\, and catalytic active site identity preclude comprehensive kinetic studies. DFT describes the electronic states of reactive intermediates and transition states but cannot access the longer length scales necessary to quantify the fluxionality of coadsorbed species or solvents. Classical simulations can accurately simulate these conformational changes but require parameterized values. Machine learning interatomic potentials have emerged as a technique to derive parameterized models from DFT data\, and our aim is to adapt these models to predict the entropy and diffusion of reactive intermediates in nanoporous catalysts.
URL:https://seasevents.nmsdev7.com/event/pics-colloquium-enabling-computationally-efficient-first-principles-kinetic-simulations-in-nanoporous-catalysts-using-machine-learning-and-data-science-with-brandon-bukowski/
LOCATION:PICS Conference Room 534 – A Wing \, 5th Floor\, 3401 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar,Colloquium
ORGANIZER;CN="Penn Institute for Computational Science (PICS)":MAILTO:dkparks@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250923T101500
DTEND;TZID=America/New_York:20250923T111500
DTSTAMP:20260403T172656
CREATED:20250902T115304Z
LAST-MODIFIED:20250902T115304Z
UID:10008487-1758622500-1758626100@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Nature in Motion: Unraveling Locomotion across Mediums and Scales"
DESCRIPTION:Biology and Engineering form an interdisciplinary two-way street. On the one side\, natural solutions can inform and inspire the design of mechanical systems. This is referred to as bioinspired design. On the other side\, referred to as engineering-enabled biology\, controlled engineering experimental\, numerical\, and analytical tools are used and developed to answer fundamental biological questions that would be difficult or even impossible to answer directly using natural systems. This talk will introduce several examples of bioinspired multifunctional structures\, such as feather-inspired flow control devices. Flow control devices on birds’ wings offer a pathway to advance the design of small aerial vehicles and inform flight control for airborne energy harvesters. In addition to bioinspired engineering\, I will highlight a few examples of engineering-enabled biology\, such as revealing the physics governing the aerial-aquatic transition of flying fish and the mechanism that enables click beetles’ legless jumping. These research topics demonstrate how nature can inform new locomotion\, actuation\, and control strategies in mechanical systems\, highlighting that engineering analysis can provide valuable insights into nature’s locomotion and adaptation strategies.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-nature-in-motion-unraveling-locomotion-across-mediums-and-scales/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250923T110000
DTEND;TZID=America/New_York:20250923T120000
DTSTAMP:20260403T172656
CREATED:20250709T150039Z
LAST-MODIFIED:20250709T150039Z
UID:10008408-1758625200-1758628800@seasevents.nmsdev7.com
SUMMARY:ESE Fall Seminar - "Architectures\, Topologies and Control for High-Frequency Power Electronics"
DESCRIPTION:Power converters designed for higher frequency operation than conventional designs can be smaller and lighter. However\, to gain these benefits and maintain high reliability the converters also need to be more efficient\, as smaller converters offer less surface area for heat removal. The development of efficient\, small and light weight power converters can benefit from converter architectures that leverage novel topologies and control techniques. Using examples from my group’s research on compact and high-efficiency power converters\, this talk will highlight the opportunities and challenges at the frontiers of high-frequency power electronics. One focus of the talk will be on new power electronic converter architectures that target high power densities and high efficiencies for wide operating range applications. Another focus of the talk will be on emerging power electronic enabled applications\, including wireless power transfer systems suitable for powering in-motion mobile platforms and radio-frequency power amplifiers for compact particle accelerators. The talk will also identify directions for future work in the area of high performance power electronic converters.
URL:https://seasevents.nmsdev7.com/event/ese-fall-seminar-title-tba-2/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Colloquium
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250923T120000
DTEND;TZID=America/New_York:20250923T130000
DTSTAMP:20260403T172656
CREATED:20250910T134335Z
LAST-MODIFIED:20250910T134335Z
UID:10008501-1758628800-1758632400@seasevents.nmsdev7.com
SUMMARY:MEAM Ph.D. Thesis Defense: "SLAM in Hard Places"
DESCRIPTION:Simultaneous Localization and Mapping is a fundamental problem for robots interacting with a novel environment and has been a densely studied area of research for several decades. The modern paradigm of feature extraction and matching coupled with advancements in sensor technology have allowed robots to achieve sub meter localization accuracy over kilometer long trajectories in controlled indoor and urban environments. Despite these advancements\, as roboticists endeavour to deploy agents in more unstructured outdoor settings to perform search and rescue or geological survey\, the standard assumptions adopted by the majority of the community start to break down. In this talk we will discuss the SLAM paradigm at a high level\, how these assumptions break down in the outdoor-unstructured setting and existing strategies for mitigation. We will present work evaluating state of the art SLAM algorithms in the geology context\, develop SLAM in an underwater ocean setting under extreme sensing constraints\, and extend this framework for active SLAM in this context.
URL:https://seasevents.nmsdev7.com/event/meam-ph-d-thesis-defense-slam-in-hard-places/
LOCATION:Eyebrow
CATEGORIES:Doctoral,Dissertation or Thesis Defense
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250923T130000
DTEND;TZID=America/New_York:20250923T150000
DTSTAMP:20260403T172656
CREATED:20250915T223943Z
LAST-MODIFIED:20250915T223943Z
UID:10008508-1758632400-1758639600@seasevents.nmsdev7.com
SUMMARY:CBE Doctoral Dissertation Defense: "DROPS-ON-STILTS MICROROBOTS FOR TRANSPORT AND DELIVERY OF HYDROPHOBIC CARGO" (Oluwafemi Ligan)
DESCRIPTION:Abstract: \nMagnetically actuated microrobots composed of dynamic assemblies of iron oxide nanoparticles (IONPs) offer versatile opportunities for biomedical applications. While such assemblies can reconfigure and perform diverse tasks under applied magnetic fields\, the effective transport of hydrophobic cargo remains a significant challenge. To address this limitation\, we introduce “drops-on-stilts” microrobots formed by the co-assembly of IONPs with oil droplets\, enabling magnetic field-guided locomotion and delivery of hydrophobic payloads\, including therapeutic agents. Using a 3D Helmholtz coil to generate uniform rotating magnetic fields\, we induce the assembly of IONPs and droplets into stilt-like architectures that lift and mobilize droplets across surfaces. The formation of these structures depends critically on particle wetting behavior\, interfacial packing\, and initial suspension conditions. At low IONP volume fractions\, magnetic dipole interactions drive IONP accumulation at droplet poles\, facilitating rotational motion. At higher concentrations\, elongated IONP chains in suspension merge with interfacially -trapped particles to create stilts\, enabling droplet walking with velocities that scale linearly with stilt length. High droplet number densities further promote chaining and merging into “log-like” collectives capable of rolling or walking. We demonstrate that drops-on-stilts and their collectives can navigate obstacles and transport both IONPs and oil as cargo. Monte Carlo simulations\, modeling IONPs as hard spheres with magnetic dipolar interactions and adhesion to droplet interfaces\, reproduce the experimentally observed chain formation and collective dynamics\, validating the proposed mechanism. Modulation of magnetic field conditions and pH tunes nanoparticle adhesion to the oil interface\, yielding distinct structural and dynamical behaviors. Finally\, we apply this platform to the delivery of hydrophobic therapeutics against Candida albicans biofilms\, highlighting its potential for biomedical treatment. Overall\, this work establishes a tunable microrobotic strategy with enhanced adaptability to complex environments\, offering a promising route for magnetically controlled transport and delivery of hydrophobic agents. \nZoom:\nMeeting ID: 819 155 7123\nPasscode: 451228
URL:https://seasevents.nmsdev7.com/event/cbe-doctoral-dissertation-defense-drops-on-stilts-microrobots-for-transport-and-delivery-of-hydrophobic-cargo-oluwafemi-ligan/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Doctoral,Student,Dissertation or Thesis Defense
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250924T120000
DTEND;TZID=America/New_York:20250924T131500
DTSTAMP:20260403T172656
CREATED:20250821T203123Z
LAST-MODIFIED:20250821T203123Z
UID:10008447-1758715200-1758719700@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "How do LLMs generalize on out-of-distribution tasks? insights from model's internal representations"
DESCRIPTION:A mystery of large language models (LLMs) is their ability to solve novel tasks\, notably through a few demonstrations in the prompt (in-context learning). Such tasks often require the model to generalize far beyond its training distribution\, raising the question: how do LLMs achieve this form of out-of-distribution (OOD) generalization? For example\, in symbolized language reasoning where names/labels are replaced by arbitrary symbols\, yet the model can infer the correct name-label mapping without any finetuning. \nIn this talk\, I will open the black box of LLMs and reveal how three facets of LLM behavior are interconnected: emergent phenomena during training\, OOD generalization\, and a model’s representation of compositions. Focusing on induction heads\, I will show that learning the right compositional structure is a key to OOD generalization\, and this learning process exhibits sharp transitions in training dynamics. Further\, I propose that “”common bridge representation hypothesis””—where a latent subspace in the embedding space acts as a bridge to align multiple attention heads across early and later layers—may be the key geometric structure underlying the success of transformers. \n  \nZoom: https://upenn.zoom.us/j/91447341103
URL:https://seasevents.nmsdev7.com/event/asset-seminar-title-tbd-3/
LOCATION:Amy Gutmann Hall\, Room 414\, 3333 Chestnut Street\, Philadelphia\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="AI-enabled Systems%3A Safe%2C Explainable%2C and Trustworthy (ASSET) Center":MAILTO:asset-info@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250924T150000
DTEND;TZID=America/New_York:20250924T160000
DTSTAMP:20260403T172656
CREATED:20250910T170329Z
LAST-MODIFIED:20250910T170329Z
UID:10008502-1758726000-1758729600@seasevents.nmsdev7.com
SUMMARY:Fall 2025 GRASP SFI: Deepti Ghadiyaram\, Boston University\, "Interpreting and Leveraging Generative Representations"
DESCRIPTION:This is a hybrid event with in-person attendance in Levine 307 and virtual attendance via Zoom.  \nABSTRACT\nGenerating high-quality photo-realistic visual content is a thriving area of research. In this talk\, I will focus not on the generation process\, but on understanding how rich visual semantic information is represented within generative models. Specifically\, I will present our work that uses mechanistic interpretability tools like k-sparse autoencoders to probe various layers and denoising timesteps of different diffusion architectures. Next\, I will present how to leverage interpretable concepts to control the generation process towards more safer and creative outcomes.
URL:https://seasevents.nmsdev7.com/event/fall-2025-grasp-sfi-deepti-ghadiyaram/
LOCATION:Levine 307\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="General Robotics%2C Automation%2C Sensing and Perception (GRASP) Lab":MAILTO:grasplab@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250924T153000
DTEND;TZID=America/New_York:20250924T163000
DTSTAMP:20260403T172656
CREATED:20250912T211221Z
LAST-MODIFIED:20250912T211221Z
UID:10008506-1758727800-1758731400@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Imaging Genetics: Enhancer-mediated Dynamic Gene Control in Space and Time" (Bomyi Lim\, University of Pennsylvania)
DESCRIPTION:Abstract: \nCanonical models suggest that regulatory enhancers function independently of genomic distance\, position\, and orientation to activate the target gene. Indeed\, a wide range of genomic distances and relative enhancer positions exist in endogenous contexts. However\, systematic analysis of how specific enhancer-promoter configurations affect transcriptional dynamics has been limited by technical constraints. Here\, we employ single-cell MS2/MCP-based live imaging in Drosophila embryos to systematically analyze transcriptional dynamics driven by varying enhancer-promoter configurations. Our analyses reveal that while linear enhancer-promoter distance moderately tunes transcriptional output\, downstream enhancer positioning relative to the promoter reduces mRNA output by 70%. Each configuration modulates distinct transcriptional parameters: linear distance governs initiation kinetics\, while relative enhancer positioning dictates transcriptional stability. Although the downstream enhancer positioning destabilized the enhancer-promoter interactions\, transcription factor hub properties remained comparable between two configurations\, revealing a clear dissociation between TF hub dynamics and E-P configuration-dependent transcriptional activity. These effects are consistent across varied enhancer and reporter sequences\, revealing configuration-dependent cis-regulatory element arrangement as an intrinsic mechanism for transcriptional fine-tuning. This work challenges the paradigm of configuration-independent enhancer function and establishes a framework to dissect the interplay between genome architecture and trans-acting factors.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-imaging-genetics-enhancer-mediated-dynamic-gene-control-in-space-and-time-bomyi-lim-university-of-pennsylvania/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar,Faculty
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
END:VCALENDAR