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:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20231105T060000
END:STANDARD
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
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241017T153000
DTEND;TZID=America/New_York:20241017T163000
DTSTAMP:20260403T172855
CREATED:20240627T183734Z
LAST-MODIFIED:20240627T183734Z
UID:10008000-1729179000-1729182600@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "DNA microscopy in two and three dimensions" (Joshua Weinstein\, University of Chicago)
DESCRIPTION:Lymphatic\, nervous\, and tumoral tissues\, among others\, exhibit physiology that emerges from three-dimensional interactions between genetically unique cells. Technologies capable of volumetrically imaging transcriptomes\, genotypes\, and morphologies in a single de novo measurement can provide a critical lens into the biological complexity of living systems. We present experimental and computational work to develop DNA microscopy: a modality of imaging that captures physical images of specimen genetic content using a massive distributed network of DNA molecules inside it. We demonstrate DNA microscopy in two-dimensional cell cultures and genome-wide in intact zebrafish embryos.
URL:https://seasevents.nmsdev7.com/event/be-seminar-joshua-weinstein-university-of-chicago/
LOCATION:216 Moore Building
CATEGORIES:Seminar
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241017T150000
DTEND;TZID=America/New_York:20241017T190000
DTSTAMP:20260403T172855
CREATED:20240930T180608Z
LAST-MODIFIED:20240930T180608Z
UID:10008112-1729177200-1729191600@seasevents.nmsdev7.com
SUMMARY:Responsible Innovation
DESCRIPTION:Rapid advancements in healthcare\, technology\, finance and beyond present significant opportunities and challenges. As we look to develop leading-edge solutions to provide a better future for all\, Penn Engineering and Perry World House are committed to promoting responsible innovation practices. \nJoin us for an inspiring afternoon as we launch the Responsible Innovation initiative at the Penn Washington Center in Washington\, DC. This event will bring together leaders in engineering\, academia\, industry\, and policy for an engaging dialogue on responsibly shaping the future of innovation. \nAgenda: \n3:00: Guests arrive \n3:15: Opening remarks by Vijay Kumar\, Nemirovsky Family Dean\, Penn Engineering \n3:25: Remarks by Ted Schlein\, partner at Kleiner Perkins\, Executive Chairman and Founding Partner at Ballistic Ventures\, and former chairman of the National Venture Capital Association (NVCA) \n3:40: Keynote address by Michael Horowitz\, Richard Perry Professor and Director of the Perry World House at the University of Pennsylvania\, and former Deputy Assistant Secretary of Defense for Force Development and Emerging Capabilities \n4:20: Faculty panel discussion moderated by Susan Davidson\, Weiss Professor of Computer and Information Science (CIS)\, featuring: \n\nMichael Kearns\, National Center Professor of Management & Technology\, CIS\nDan Koditschek\, Alfred Fitler Moore Professor\, Electrical and Systems Engineering (ESE)\, CIS\, and Mechanical Engineering and Applied Mechanics (MEAM)\nMark Yim\, Asa Whitney Professor of Mechanical Engineering\, MEAM\, CIS\, and ESE\n\n5:00: Presentation by George Pappas\, UPS Foundation Professor of Transportation\, ESE\, CIS\, and MEAM \n5:30: Cocktail Hour \nDon’t miss this opportunity to engage with thought leaders and be part of Penn Engineering’s bold vision for advancing responsible innovation. \nPlease make sure to RSVP by Friday\, October 11\, to secure your spot.
URL:https://seasevents.nmsdev7.com/event/responsible-innovation/
LOCATION:Penn Washington Center\, 101 Constitution Ave. NW\, Washington DC\, DC\, 20001\, United States
CATEGORIES:Faculty,Panel Discussion
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241017T140000
DTEND;TZID=America/New_York:20241017T150000
DTSTAMP:20260403T172855
CREATED:20241009T155327Z
LAST-MODIFIED:20241009T155327Z
UID:10008129-1729173600-1729177200@seasevents.nmsdev7.com
SUMMARY:Fall 2024 GRASP Seminar: Bob Collins & Yanxi Liu\, Penn State University\, “From Vision to Dynamics”
DESCRIPTION:*This seminar will be held in-person in Levine 307 as well as virtually via Zoom. \nABSTRACT\nIn the realms of health and sports\, quantitative analysis of human movement provides guidance for individualized performance characterization\, training\, and health assessment. Data collection for biomechanical analysis of human motion is typically performed using expensive\, specialized hardware that requires experiments to occur in a constrained lab setting. Computer vision algorithms using inexpensive\, off-the-shelf video cameras for data collection would enable observation in more natural environments with minimal cost. Towards this end\, this talk reports on:\n• Our effort in collecting the first publicly available Taiji-MultiModal (PSU TMM100) dataset containing 100 sequences of simplified 24-form Taiji with synchronized mocap\, video\, and foot-insole pressure maps.\n• An unsupervised learning method for training a light-weight encoder suitable for 3D body pose classification and sequence-to-sequence temporal alignment.\n• The first deep learning baseline that demonstrates reliable and repeatable mapping from either a single frame or short sequence of human pose (kinematics) into predicted foot pressure map output (dynamics)\, leading to image-based stability monitoring in natural environments.
URL:https://seasevents.nmsdev7.com/event/fall-2024-grasp-seminar-robert-collins-yanxi-liu-penn-state-university-from-vision-to-dynamics/
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:20241017T110000
DTEND;TZID=America/New_York:20241017T120000
DTSTAMP:20260403T172855
CREATED:20241010T181407Z
LAST-MODIFIED:20241010T181407Z
UID:10008131-1729162800-1729166400@seasevents.nmsdev7.com
SUMMARY:IDEAS Seminar: "Equivariant Neural Inertial Odometry"
DESCRIPTION:Abstract:  \nIn this talk\, we introduce a new class of problems related to integrating inertial measurements obtained from an IMU that play a significant role in navigation combined with visual data. While there have been tremendous technological advances in the precision of instrumentation\, integrating acceleration and angular velocity still suffers from drift in the displacement estimates. Neural networks have come to the rescue in estimating displacement and the associated uncertainty covariance. However\, such networks do not consider the physical roto-reflective symmetries inherent in IMU data\, leading to the need to memorize the same priors for every possible motion direction\, which hinders generalization. In this work\, we characterize these symmetries and show that the IMU data and the resulting displacement and covariance transform equivariantly when rotated around the gravity vector and reflected with respect to arbitrary\nplanes parallel to gravity. We propose a network for predicting an equivariant gravity-aligned frame from equivariant vectors and invariant scalars derived from IMU data\, leveraging expressive linear and non-linear layers tailored to commute with the underlying symmetry transformation. Such a canonical frame can precede existing architectures that are end-to-end or filter-based. We will include an introduction to the inertial filtering problem and we will present\nresults in real-world datasets. \nZoom link: https://caltech.zoom.us/j/88141815847?pwd=ayPBbwoDE91IWysv1P4Oxp6zbamrSQ.1
URL:https://seasevents.nmsdev7.com/event/ideas-seminar-equivariant-neural-inertial-odometry/
LOCATION:Room 401B\, 3401 Walnut\, 3401 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar,Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241017T103000
DTEND;TZID=America/New_York:20241017T120000
DTSTAMP:20260403T172855
CREATED:20241008T150959Z
LAST-MODIFIED:20241008T150959Z
UID:10008128-1729161000-1729166400@seasevents.nmsdev7.com
SUMMARY:MSE Seminar “Shaping a Soft Future"
DESCRIPTION:Existing devices—such as cell phones\, computers\, and robots – are made from rigid materials\, which is in direct contrast to the soft materials that compose the human body. In this talk\, I will discuss several topics related to studying and harnessing soft materials within the context of creating devices with tissue like properties. \n\nLiquid metal: Gallium-based liquid metals are often overlooked despite their remarkable properties: melting points below room temperature\, water-like viscosity\, low-toxicity\, and effectively zero vapor pressure (they do not evaporate). They also have\, by far\, the largest interfacial tension of any liquid at room temperature. Normally small volumes of liquids with large tension form spherical or hemi-spherical structures to minimize surface energy. Yet\, these liquid metals can be patterned into non-spherical shapes (cones\, wires\, antennas) due to a thin\, oxide skin that forms rapidly on its surface.\nShape reconfiguration: Perhaps the most fascinating aspect of liquid metals it the ability to use interfacial electrochemistry chemistry to remove / deposit the oxide to manipulate the surface tension of the metal over unprecedented ranges (from the largest tension of any known liquid to near zero!). This allows manipulating the shape and position of the metal for shape reconfigurable devices.\nLonogels: Soft materials that are tough (that is\, they do not readily tear or fail mechanically) are important for a number of applications\, including encapsulation of devices. Recently\, we discovered a simple way to create ultra-tough ionogels\, which are polymer networks swollen with ionic liquids. These materials are tougher than cartilage and compatible with 3D printing.\n\nThis work has implications for soft and stretchable electronics; that is\, devices with desirable mechanical properties for human-machine interfacing\, soft robotics\, and wearable electronics.
URL:https://seasevents.nmsdev7.com/event/mse-seminar-shaping-a-soft-future/
LOCATION:Wu & Chen Auditorium
CATEGORIES:Seminar
ORGANIZER;CN="Materials Science and Engineering":MAILTO:johnruss@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241016T153000
DTEND;TZID=America/New_York:20241016T163000
DTSTAMP:20260403T172855
CREATED:20240816T202836Z
LAST-MODIFIED:20240816T202836Z
UID:10008053-1729092600-1729096200@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Optimizing Work and Heat flows in Sustainable Chemical Processes Using Attainable Regions" (Diane Hildebrandt\, Rutgers University)
DESCRIPTION:Abstract: \nA chemical process transforms feed materials into desired products\, necessitating the exchange of heat and work between the process and the environment. Optimizing process performance requires careful selection of chemical pathways\, process parameters\, and heat/workflows. The reversible process sets the upper-performance limit for sustainable design\, thereby setting a target for process design. We propose using a plot of enthalpy (H) against Gibbs Free Energy (G) to represent\, synthesize\, and analyze heat and workflows in chemical processes. This graphical approach is particularly useful for identifying and assessing process reversibility. We define the Thermodynamic Attainable Region (AR T ) as the set of all (delta)H and (delta)G across all possible processes that convert a given feed(s) into specified products. In this talk\, we will present a methodology for calculating the AR T . By leveraging the visual representation of the AR T \, we can systematically analyze and compare the inherent heat and workflows of different processes\, feeds\, and products. We will consider some simple examples to illustrate the impact of the selection of chemical pathways\, process parameters\, and feeds on process heat and workflows and reversibility.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-optimizing-work-and-heat-flows-in-sustainable-chemical-processes-using-attainable-regions-diane-hildebrandt-rutgers-university/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241016T150000
DTEND;TZID=America/New_York:20241016T160000
DTSTAMP:20260403T172855
CREATED:20241010T152537Z
LAST-MODIFIED:20241010T152537Z
UID:10008130-1729090800-1729094400@seasevents.nmsdev7.com
SUMMARY:Fall 2024 GRASP SFI: Jun-Yan Zhu\, Carnegie Mellon University\, "Ensuring Data Ownership in Generative Visual Models"
DESCRIPTION:This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. \nABSTRACT\nLarge-scale generative visual models have made content creation as little effort as writing a short text description. However\, these models are typically trained on an enormous amount of Internet data\, often containing copyrighted material\, licensed images\, and personal photos. How can we remove these images if creators decide to opt out? How can we properly compensate them if they choose to opt in? \nIn this talk\, I will first describe an efficient method for removing copyrighted materials\, artistic styles of living artists\, and memorized images from pretrained text-to-image models. I will then discuss our data attribution algorithms for assessing the influence of each training image for a generated sample. Collectively\, we aim to enable creators to retain control over the ownership of training images.
URL:https://seasevents.nmsdev7.com/event/fall-2024-grasp-sfi-jun-yan-zhu/
LOCATION:Levine 307\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
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:20241016T133000
DTEND;TZID=America/New_York:20241016T143000
DTSTAMP:20260403T172855
CREATED:20241004T153846Z
LAST-MODIFIED:20241004T153846Z
UID:10008121-1729085400-1729089000@seasevents.nmsdev7.com
SUMMARY:MEAM Ph.D. Thesis Defense: "Semantics-Driven Active Perception and Navigation with Aerial Robots"
DESCRIPTION:Autonomous aerial robots today are capable of safely navigating through cluttered\, GPS-denied environments while constructing an accurate map that captures geometric features such as points\, lines\, and planes. Such maps are crucial for low-level planning and obstacle avoidance. However\, beyond offering details on the density\, layout\, and dimensions of the environment\, these maps provide limited information for semantically meaningful reasoning. For instance\, they fall short in helping the robot identify where to find specific objects during search and rescue operations\, which areas are relevant during infrastructure inspection or asset mapping\, or in distinguishing between static and dynamic entities in the environment during localization and navigation. Such reasoning capabilities are especially important for resource-constrained robots\, such as micro aerial vehicles\, deployed in large-scale environments. In such scenarios\, robots need to keep track of important\, actionable information\, while respecting their onboard computational and storage resource constraints. Sparse\, semantically meaningful maps facilitate robust state estimation and storage-efficient mapping\, while also enabling high-level reasoning that guides intelligent\, task-relevant decision-making during navigation and exploration. \nThis thesis introduces a set of novel methodologies and algorithms for semantics-driven perception and autonomy\, enabling robots to safely navigate and explore large-scale\, complex environments while actively and collaboratively constructing high-quality\, semantically meaningful maps. Specifically\, we first present a monocular-camera-based semantic mapping system that integrates deep learning\, visual tracking\, and semantic Structure from Motion (SfM) for accurate fruit detection and mapping in orchards. We then develop a semantic Simultaneous Localization and Mapping (SLAM) integrated autonomous aerial navigation system for large-scale semantic mapping in under-forest-canopy environments. This system leverages real-time semantic SLAM for accurate pose estimation and timber metric assessment. Next\, we introduce active metric-semantic SLAM systems for both urban outdoor and indoor environments. These systems use semantics to guide aerial robots to explore the environment and minimize both metric and semantic uncertainties. Lastly\, bridging all efforts together\, we propose a decentralized metric-semantic SLAM framework for autonomous navigation and exploration with heterogeneous robot teams operating across multiple environments. Extensive real-world experiments have validated the robustness and performance of the proposed methods across multiple aerial and ground robot platforms in various environments\, including multi-floor indoor spaces\, urban outdoors\, forests\, and orchards. Field deployments further showcase the system’s potential for direct application in solving important real-world problems\, such as precision agriculture\, forestry management\, climate change mitigation\, infrastructure inspection\, and factory asset management. Finally\, the proposed systems and algorithms are made available as open-source tools for public use.
URL:https://seasevents.nmsdev7.com/event/meam-ph-d-thesis-defense-semantics-driven-active-perception-and-navigation-with-aerial-robots/
LOCATION:Room 313\, Singh Center for Nanotechnology\, 3205 Walnut Street\, Philadelphia\, PA\, 19104\, United States
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:20241016T120000
DTEND;TZID=America/New_York:20241016T131500
DTSTAMP:20260403T172855
CREATED:20240715T210338Z
LAST-MODIFIED:20240715T210338Z
UID:10008021-1729080000-1729084500@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "Some Displaced Vignettes on Generalized Notions of Equivariance"
DESCRIPTION:Abstract: \nThe explicit incorporation of task-specific inductive biases through symmetry has emerged as a crucial design precept in the development of high-performance machine learning models. Symmetry-aware neural networks\, such as group equivariant networks\, have achieved notable success in areas like protein and drug design\, where capturing task-specific symmetries improves generalization. Recent efforts have focused on models that relax equivariance\, balancing flexibility and equivariance to enhance performance. In the first part of the talk\, I will discuss the benefits of partial and approximate equivariance from a theoretical perspective\, presenting quantitative bounds that demonstrate how models capturing task-specific symmetries lead to improved generalization. Utilizing this quantification\, I will examine the more general question of dealing with approximate/partial symmetries and model mis-spefication\, delineating conditions under which the model equivariance is optimal for a given level of data symmetry. In the second part\, I will present a general formalism based on special structured matrices\, which generalizes classical low-displacement rank theory of Kailath and co-workers\, which can help in constructing approximately equivariant neural networks with significantly reduced parameter counts. In the last part\, I will discuss some attempts at generalizing notions of equivariance in the context of language and compositional generalization. I will also talk about some ongoing work on using such notions for the problem of inverse protein folding. \nWork done in collaboration with: Mircea Petrace (Pontificia Universidad Católica de Chile)\, Ashwin Samudre (Simon Fraser University)\, Brian D. Nord (Fermilab and University of Chicago)\, and Payel Das (IBM Research). \nZoom Link (if unable to attend in-person): https://upenn.zoom.us/j/96014696752
URL:https://seasevents.nmsdev7.com/event/asset-seminar-shubhendu-trivedi-massachusetts-institute-of-technology/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241016T110000
DTEND;TZID=America/New_York:20241016T120000
DTSTAMP:20260403T172855
CREATED:20241003T134027Z
LAST-MODIFIED:20241003T134027Z
UID:10008118-1729076400-1729080000@seasevents.nmsdev7.com
SUMMARY:ESE Guest Seminar - "Safe Offline RL for Constrained Markov Decision Process: Theory and Practice"
DESCRIPTION:Many constrained sequential decision-making processes such as safe AV navigation\, wireless network control\, caching\, cloud computing\, etc.\, can be cast as Constrained Markov Decision Processes (CMDP). Reinforcement Learning (RL) algorithms have been used to learn optimal policies for unknown unconstrained MDP. Extending these RL algorithms to unknown CMDP\, brings the additional challenge of not only maximizing the reward but also satisfying the constraints. Further\, in most of the practical applications\, one has to rely on the offline database as online interaction might be costly or infeasible. \nWhile the unconstrained offline RL setting is relatively well-understood\, the offline CMDP or safe offline RL setup is not. For example\, consider a database that consists of data coming from a safe behavioral policy\, it remained an open problem on how to develop an algorithm that would provide safety while maximizing the reward with provable guarantee. In particular\, the existing works on safe offline RL rely on the assumption that the database must contain state-action pairs coming from all the policies which is not practical in safety-critical setup as the database might not contain unsafe state-action pairs. We closed the gap in our recent research. In our work\, we developed a weighted safe actor-critic (WSAC) algorithm that can produce a policy that outperforms any behavioral policy while maintaining the same level of safety\, which is critical to designing a safe algorithm for offline RL. Additionally\, we compare WSAC with existing state-of-the-art safe offline RL algorithms in several continuous control environments. WSAC outperforms all baselines across a range of tasks\, supporting the theoretical results.
URL:https://seasevents.nmsdev7.com/event/ese-seminar-safe-offline-rl-for-constrained-markov-decision-process-theory-and-practice/
LOCATION:Greenberg Lounge (Room 114)\, Skirkanich Hall\, 210 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241015T153000
DTEND;TZID=America/New_York:20241015T163000
DTSTAMP:20260403T172855
CREATED:20241003T164247Z
LAST-MODIFIED:20241003T164247Z
UID:10008119-1729006200-1729009800@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Optimal Oblivious Reconfigurable Networks"
DESCRIPTION:As Moore’s Law slows down\, packet switch capabilities are falling behind datacenter demands. Recent hardware advances have enabled the new switching technology of nanosecond-scale rapid circuit switches. Combined with novel network designs\, these have the potential to fully replace packet switches. This talk presents the Oblivious Reconfigurable Network (ORN) design paradigm which is ideally suited to this new switching technology. I describe how to design ORNs that work at datacenter scale\, supporting tens of thousands of network nodes. And\, I discuss an implementation\, Shale\, whose tradeoffs in latency and throughput are Pareto optimal among all ORN designs achieving orders of magnitude better latency and memory requirements than prior ORN designs at such scales. \nA paper of this works appears in SIGCOMM 2024: “Shale: A Practical\, Scalable Oblivious Reconfigural Networks”
URL:https://seasevents.nmsdev7.com/event/cis-seminar-optimal-oblivious-reconfigurable-networks/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241015T110000
DTEND;TZID=America/New_York:20241015T120000
DTSTAMP:20260403T172855
CREATED:20240821T140955Z
LAST-MODIFIED:20240821T140955Z
UID:10008065-1728990000-1728993600@seasevents.nmsdev7.com
SUMMARY:ESE Fall Seminar - "Learning-NUM: Utility Maximization in Stochastic Queueing Networks"
DESCRIPTION:We consider the problem of network utility maximization (NUM) and propose a new Learning-NUM framework\,  where the users’ utility functions are unknown apriori and the utility function values can be observed only after the corresponding traffic is delivered to the destination.   We start by considering linear utility functions and propose a priority-based network control policy\, that combines techniques from network control and multi-arm bandits to achieve logarithmic regret.  We then consider the case of concave utility functions and design the Gradient Sampling Max-Weight algorithm (GSMW)\, based on the ideas of gradient estimation and Max-Weight scheduling\, that achieves sublinear utility regret.  We further demonstrate the applicability of the gradient sampling approach to minimum delay routing in wireless networks. Finally\, we consider the general problem of reinforcement learning for queueing networks with unbounded state-spaces\, with the goal of making control decisions that minimizing the queue length.  We formulate the problem as an MDP\, and propose a new reinforcement learning framework\, called Truncated Upper Confidence Reinforcement Learning (TUCRL)\, that  can achieve optimal performance.  We show how this framework can be applied to deep reinforcement learning (DRL) for online stochastic network optimization.
URL:https://seasevents.nmsdev7.com/event/ese-fall-seminar-title-tbd-16/
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:20241015T101500
DTEND;TZID=America/New_York:20241015T111500
DTSTAMP:20260403T172855
CREATED:20240828T133408Z
LAST-MODIFIED:20240828T133408Z
UID:10008075-1728987300-1728990900@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Propulsive Advantages of Coordinating Multiple Jets by Colonial Marine Organisms"
DESCRIPTION:Salps and siphonophores are widespread marine animals that occur in centimeters to meters-long colonial chains and employ multiple\, pulsed swimming jets. We use a combination of approaches including in situ and lab experiments\, flow visualizations and custom high-speed camera systems to understand the morphology\, kinematics and fluid mechanics that underpin efficient movement in these colonial chains. Our results over the past decade show that: 1) kinematics drive effective fluid refill and maneuverability\, 2) coordination among swimming units produces different swimming modes and enhances efficiency and\, 3) diverse colony architectures give rise to a range of performance solutions. Colonial salps and siphonophores are a model system for understanding structure-fluid interactions in multi-jet locomotion. Our studies using 3D videography in the ocean will hopefully further our understanding of coordinated pulsed jets\, and elucidate design principles optimized by nature.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-propulsive-advantages-of-coordinating-multiple-jets-by-colonial-marine-organisms/
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:20241011T140000
DTEND;TZID=America/New_York:20241011T150000
DTSTAMP:20260403T172855
CREATED:20240828T190959Z
LAST-MODIFIED:20240828T190959Z
UID:10008078-1728655200-1728658800@seasevents.nmsdev7.com
SUMMARY:PICS Colloquium: Mean flow and turbulence in unsteady urban canopy flows
DESCRIPTION:Advancing the current understanding and capability to predict atmospheric flow and related transport in urban areas is critical for many applications\, including air quality modeling\, urban climate\, pedestrian comfort and structural resilience. Turbulence in these environments is rarely in equilibrium with the underlying surface and is typically characterized by strong departures from statistical stationarity. For example\, the atmospheric boundary is often driven by a range of (sub)meso forcings that can evolve over sufficiently short time scales and result in unsteady flow conditions in cities. Yet\, current theories describing mass\, energy\, and momentum transport in urban areas are largely established for equilibrium and statistically stationary flow. This presentation will provide an overview on the effects of a particular class of flow unsteadiness (flow pulsation) on the structure of mean flow and turbulence in urban areas. The discussion will focus on the impact of flow unsteadiness on flow statistics and on the geometry of coherent structures responsible for momentum and kinetic energy transport.
URL:https://seasevents.nmsdev7.com/event/pics-mean-flow-and-turbulence-in-unsteady-urban-canopy-flows/
LOCATION:PICS Conference Room 534 – A Wing \, 5th Floor\, 3401 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Colloquium
ORGANIZER;CN="Penn Institute for Computational Science (PICS)":MAILTO:dkparks@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241011T103000
DTEND;TZID=America/New_York:20241011T114500
DTSTAMP:20260403T172855
CREATED:20241004T194240Z
LAST-MODIFIED:20241004T194240Z
UID:10008120-1728642600-1728647100@seasevents.nmsdev7.com
SUMMARY:Fall 2024 GRASP on Robotics: Antonio Loquercio\, University of Pennsylvania\, "Simulation: What made us intelligent will make our robots intelligent"
DESCRIPTION:This will be a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. \nABSTRACT\nSimulation-to-reality transfer is an emerging approach that enables robots to develop skills in simulated environments before applying them in the real world. This method has catalyzed numerous advancements in robotic learning\, from locomotion to agile flight. In this talk\, I will explore simulation-to-reality transfer through the lens of evolutionary biology\, drawing intriguing parallels with the function of the mammalian neocortex. By reframing this technique in the context of biological evolution\, we can uncover novel research questions and explore how simulation-to-reality transfer can evolve from an empirically driven process to a scientific discipline.
URL:https://seasevents.nmsdev7.com/event/fall-2024-grasp-on-robotics-antonio-loquercio-university-of-pennsylvania-simulation-what-made-us-intelligent-will-make-our-robots-intelligent/
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:20241010T153000
DTEND;TZID=America/New_York:20241010T163000
DTSTAMP:20260403T172855
CREATED:20240923T162421Z
LAST-MODIFIED:20240923T162421Z
UID:10008104-1728574200-1728577800@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Neurosymbolic AI for Safety-Critical Agile Control"
DESCRIPTION:This talk overviews research at Caltech on designing hybrid or neurosymbolic AI systems that blend learning with symbolic structure\, in order to achieve both the flexibility of the former and the formal interpretability and generalization power of the latter.  By having systems that are formally interpretable\, one can employ a wide range of formal analysis techniques to verify important properties of the overall system\, such as those related to safety and stability\, as well as use those analyses to guide system design and optimization.  Focusing on formally interpretable structure arising from control and planning\, I will present new algorithms and their deployment in a range of applications\, including agile flight control under challenging and time-varying environments\, controlling highly underactuated systems (e.g.\, one-legged hoppers)\, as well as briefly overview other related research.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-neurosymbolic-ai-for-safety-critical-agile-control-2/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241010T153000
DTEND;TZID=America/New_York:20241010T163000
DTSTAMP:20260403T172855
CREATED:20240627T183240Z
LAST-MODIFIED:20240627T183240Z
UID:10007999-1728574200-1728577800@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "Synthetic reconstitution of complex cellular behavior" (Ahmad Khalil\, Boston University)
DESCRIPTION:Cells use genetically-encoded molecular circuits to execute diverse biological functions. We are developing novel tools of synthetic biology that allow us to construct regulatory circuitry inside living cells that recapitulate complex functions like those seen in nature. In this talk\, I will describe how we use this approach to achieve three objectives. First\, I will demonstrate how synthetic reconstitution provides a powerful way to understand fundamental principles of regulatory networks\, which we have applied to guide discoveries in eukaryotic transcription regulation and epigenetics. Second\, I will describe our advancements in developing synthetic circuit technologies that enable precise\, instructive control of therapeutic human cell function to address challenges in emerging cellular therapies\, such as CAR-T cells for cancer. Finally\, I will share a future vision of how synthetic reconstitution can be used to engineer cells with little or no intrinsic therapeutic potential into powerful and scalable engines for generating custom\, therapeutically-relevant molecules. Overall\, by learning how to build biological systems from scratch\, our broad goal is to connect the basic molecular building blocks of life to complex cellular behavior and ultimately to clinical applications.
URL:https://seasevents.nmsdev7.com/event/be-seminar-ahmad-khalil-boston-university/
LOCATION:216 Moore Building
CATEGORIES:Seminar
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241010T110000
DTEND;TZID=America/New_York:20241010T120000
DTSTAMP:20260403T172855
CREATED:20240821T140706Z
LAST-MODIFIED:20240821T140706Z
UID:10008064-1728558000-1728561600@seasevents.nmsdev7.com
SUMMARY:ESE Fall Seminar - "Deep Latent Variable Models for Compression and Natural Science"
DESCRIPTION:Latent variable models have been an integral part of probabilistic machine learning\, ranging from simple mixture models to variational autoencoders to powerful diffusion probabilistic models at the center of recent media attention. Perhaps less well-appreciated is the intimate connection between latent variable models and data compression\, and the potential of these models for advancing natural science. This talk will explore these topics. I will begin by showcasing connections between variational methods and the theory and practice of neural data compression. On the applied side\, variational methods lead to machine-learned compressors of data such as images and videos and offer principled techniques for enhancing their compression performance\, as well as reducing their decoding complexity. On the theory side\, variational methods also provide scalable bounds on the fundamental compressibility of real-world data\, such as images and particle physics data. Lastly\, I will also delve into applications\, where I show how deep latent variable models allow solving challenging inverse problems in weather and climate modeling tasks.
URL:https://seasevents.nmsdev7.com/event/ese-fall-seminar-title-tbd-15/
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:20241010T103000
DTEND;TZID=America/New_York:20241010T120000
DTSTAMP:20260403T172855
CREATED:20240923T150045Z
LAST-MODIFIED:20240923T150045Z
UID:10008103-1728556200-1728561600@seasevents.nmsdev7.com
SUMMARY:MSE Seminar: "Color From Colorless Materials: Harnessing Multi-reflection Interference in Microstructures"
DESCRIPTION:Many of the colors found in nature\, such as those of iridescent\, color-shifting organisms like beetles\, butterflies\, and birds\, are structural colors. Structural coloration is often generated by optical interference occurring within nanoscale periodic structures\, like diffraction gratings\, photonic crystals\, or thin films. In these cases\, the periodicity of the structure is similar to the wavelength of the visible light undergoing interference. However\, I will describe the interesting observation and mechanism behind how optical interference and iridescent color can be generated by light interacting within much larger\, microstructure structures. In this mechanism\, light reflecting multiple times (such as by total internal reflection or a mirrored surface) and traveling along different paths within a microstructure can optically interfere. This effect happens in materials as simple as water droplets\, but it can also be harnessed within far more complex 3D polymeric geometries to customize the interference. Ray tracing simulations coupled with color visualization and spectral analysis techniques can be used to model\, examine\, and rationalize the iridescence generated for a range of micro-geometries\, including hemicylinders\, hemispheres\, truncated hemispheres\, and other irregular structures under varying illumination conditions. Microstructure arrays patterned on surfaces with varying orientation and size lead to unique color-traveling optical effects and highlight opportunities for how multibounce reflection interference can be used to create customizable colored appearances. The findings provide a conceptual framework for rationalizing the multibounce interference mechanism and establish approaches for characterizing and tailoring the optical and iridescent properties of microstructured surfaces.
URL:https://seasevents.nmsdev7.com/event/mse-seminar-color-from-colorless-materials-harnessing-multi-reflection-interference-in-microstructures/
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:20241009T153000
DTEND;TZID=America/New_York:20241009T163000
DTSTAMP:20260403T172855
CREATED:20240816T202436Z
LAST-MODIFIED:20240816T202436Z
UID:10008052-1728487800-1728491400@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Organic Batteries for a More Sustainable Future" (Jodie Lutkenhaus\, Texas A&M University)
DESCRIPTION:Abstract: \nCobalt\, nickel\, and lithium are essential ingredients in today’s lithium-ion batteries (LIBs)\, but their continued use presents economic\, ethical\, and environmental challenges. Society must now begin to consider the implications of a LIB’s full life cycle\, including the carbon footprint\, the economic and environmental costs\, and material access. These challenges motivate the case for degradable or recyclable batteries sourced from earth-abundant materials whose life cycle bears minimal impact on the environment. This presentation considers organic polymer-based batteries\, which have the potential to address many of these issues. Redox-active polymers form the positive and negative electrodes\, storing charge through a reversible redox mechanism. We demonstrate polypeptide radical batteries that degrade on command into amino acids and by-products as a first step toward circular organic batteries. Further\, we show the recycling of redox-active polymer electrodes using a solvent-based approach. Polymer-air batteries are examined as high-capacity alternatives to metal-air batteries. The molecular mechanism for each case is investigated\, revealing pathways forward for improving each polymer’s performance. Taken together\, organic batteries offer the promise of a circular platform free of critical elements.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-organic-batteries-for-a-more-sustainable-future-jodie-lutkenhaus-texas-am-university/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241009T150000
DTEND;TZID=America/New_York:20241009T160000
DTSTAMP:20260403T172855
CREATED:20241004T191650Z
LAST-MODIFIED:20241004T191650Z
UID:10008122-1728486000-1728489600@seasevents.nmsdev7.com
SUMMARY:Fall 2024 GRASP SFI: Tushar Kusnur\, The AI Institute\, “Robotic Information Gathering: Experiences and Perspectives”
DESCRIPTION:This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. \nABSTRACT\nDespite all the interest and hype both from within and outside our community\, robotics is a relatively new and highly specialized field. We embody substantial diversity in research direction and style\, the focus and flavor of industrial companies\, and academic departments and programs. There is no single way to “do” robotics. Similarly\, the journeys of any two people in this field are unlikely to look the same. My work so far has shared a theme of “active information gathering” in academic research for novel approaches in specific contexts\, as well as industry research that balances novel and practical systems in broader contexts. Consistent with that distinction\, I will first talk about my past work at Carnegie Mellon including planning for robotic coverage and active sensing grounded in a system of multiple aerial vehicles. Next\, I will briefly highlight some technical challenges relevant to my current work on the “Watch\, Understand\, Do” project at the AI Institute\, grounded in mobile manipulation. I will share what I have learned about how different styles of research\, collaboration\, and project planning lend themselves to these different settings\, with an eye on how we can make more informed decisions in our individual and collective journeys as roboticists.
URL:https://seasevents.nmsdev7.com/event/fall-2024-grasp-sfi-tushar-kusnur/
LOCATION:Levine 307\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
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:20241009T120000
DTEND;TZID=America/New_York:20241009T131500
DTSTAMP:20260403T172855
CREATED:20240709T173859Z
LAST-MODIFIED:20240709T173859Z
UID:10008011-1728475200-1728479700@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "Wood Wide Models"
DESCRIPTION:Abstract:  \nFoundation models are monolithic models that are trained on a broad set of data\, and which are then in principle fine-tuned to various specific tasks. But they are ill-suited to many heterogeneous settings\, for instance numeric tabular data\, or numeric time-series data\, where training a single monolithic model over a large collection of such datasets is not meaningful. For instance\, why should numeric times series of stock prices have anything to do with time series comprising the vital signs of an ICU patient? For such settings\, we propose the class of wood wide models. \nThe wood wide web is often used to describe an underground network of fungal threads that connect many trees and plants together\, which stands in contrast to a large concrete foundation on top of which we might build specialized buildings. Analogously\, in contrast to a single foundation model upon which one might build specialized models\, we can have many smaller wood wide models that all borrow subtler ingredients from each other. But to be able to share nutrients from the wood wide web\, trees need a special root based architecture that can connect to these fungal threads. Accordingly\, to operationalize wood wide models\, we develop a novel neuro-symbolic architecture\, that we term “neuro-causal”\,  that uses a synthesis of deep neural models and causal graphical models to automatically infer higher level symbolic information from lower level “raw features”\, while also allowing for rich relationships among the symbolic variables. Neuro-causal models retain the flexibility of modern deep neural network architectures while simultaneously capturing statistical semantics such as identifiability and causality\, which are important to discuss ideal\, target representations and their tradeoffs. But most interestingly\, these can further form a web of wood wide models when they borrow in part from a shared conceptual ontology\, as well as causal mechanisms. We provide conditions under which this entire architecture can be recovered uniquely. We also discuss efficient algorithms and provide experiments illustrating the algorithms in practice. \nZoom Link (if unable to attend in-person): https://upenn.zoom.us/j/98029108883
URL:https://seasevents.nmsdev7.com/event/asset-seminar-pradeep-ravikumar-carnegie-melon-university/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241009T090000
DTEND;TZID=America/New_York:20241009T110000
DTSTAMP:20260403T172855
CREATED:20240925T180900Z
LAST-MODIFIED:20240925T180900Z
UID:10008107-1728464400-1728471600@seasevents.nmsdev7.com
SUMMARY:CBE Doctoral Dissertation Defense: "Investigation of Thin-film Cerium-based Oxides and Single Atom Catalysts Prepared by Atomic Layer Deposition" (Kai Shen)
DESCRIPTION:Abstract:\nMetal oxides play a critical role in heterogeneous catalysis\, acting as active catalysts in a wide range of reactions from dehydrogenation to selective and complete oxidation. However\, many catalytically active metal oxides suffer from low surface areas in their bulk form\, particularly after high-temperature calcination. This thesis focuses on cerium-based oxides due to their exceptional catalytic properties\, particularly in oxygen storage and release. By synthesizing these materials as supported thin films on high surface area γ-Al2O3\, we demonstrate that CeFeOx\, CeMnOx\, and CeVOx thin films not only retain high surface areas under elevated temperatures but also exhibit unique properties compared to their bulk counterparts\, including differences in lattice structure and redox behavior. When employed as a support for precious metals like Pt\, the choice of reducing or oxidizing pretreatment significantly impacts catalytic activity\, as it modulates the interaction between the support and metal. Furthermore\, we extended the ALD technique to the synthesis of single-atom catalysts. We invented a novel method\, Diluted Atomic Layer Deposition (DALD)\, to fabricate single-atom catalysts with precisely tunable weight loadings. DALD involves using a mixture of metal precursors and non-metallic ligands (e.g.\, Pt(acac)2 and Hacac) in controlled ratios. We demonstrated the successful deposition of Ir\, Rh\, and Pt onto γ-Al2O3\, achieving weight loadings ranging from 4 wt% down to as low as 0.04 wt%\, with corresponding metal particle sizes ranging from ~2 nm down to isolated single atoms. \nZoom Information:\nMeeting ID: 397 149 8817\nPasscode: uuVkZ2
URL:https://seasevents.nmsdev7.com/event/cbe-doctoral-dissertation-defense-investigation-of-thin-film-cerium-based-oxides-and-single-atom-catalysts-prepared-by-atomic-layer-deposition-kai-shen/
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:20241008T153000
DTEND;TZID=America/New_York:20241008T163000
DTSTAMP:20260403T172855
CREATED:20240927T162100Z
LAST-MODIFIED:20240927T162100Z
UID:10008110-1728401400-1728405000@seasevents.nmsdev7.com
SUMMARY:CIS Special Industry Seminar: "Improving Foundation Models Using Human Data"
DESCRIPTION:Foundation models including LLMs and multi-modal models released by OpenAI (GPT)\, Anthropic (Claude)\, Google (Gemini)\, Meta (Llama)\, and others have shown very impressive capabilities across a range of tasks. Some key drivers of this performance — such as investments in GPUs/compute\, model size\, and pre-training data — are relatively well understood. \nThis presentation will focus on a less understood\, yet extremely powerful lever that creates significant differentiation and competitive advantage among state-of-the-art models: the use of expert human data for Evaluations (“Evals”)\, Supervised Fine Tuning (“SFT”)\, Reinforcement Learning with Human Feedback (“RLHF”)\, and Direct Preference Optimization (“DPO”). \nThe talk will also outline some best practices for maximizing returns on financial investments in human data to achieve optimal model performance. This includes effective strategies for sourcing\, vetting\, hiring\, and managing expert human data teams\, as well as task design for Evals\, SFT\, RLHF\, and DPO\, along with processes and tooling to optimize team performance\, data quality and throughput. \nThis talk is open to Penn students\, COLM attendees\, and anyone in the Philadelphia area interested in AI\, data science\, or machine learning. You don’t need to be affiliated with the conference or the university — if you’re part of the DS/ML/AI community\, you’re welcome to join us.
URL:https://seasevents.nmsdev7.com/event/cis-special-industry-seminar-improving-foundation-models-using-human-data/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241008T110000
DTEND;TZID=America/New_York:20241008T120000
DTSTAMP:20260403T172855
CREATED:20240821T140330Z
LAST-MODIFIED:20240821T140330Z
UID:10008063-1728385200-1728388800@seasevents.nmsdev7.com
SUMMARY:ESE Fall Seminar - "Reflections on learning about learning: A case study on where ideas come from in (In-)Secure Processor Design"
DESCRIPTION:When it comes to security\, hardware is the new software. Starting some years ago\, this shift was made plain when a litany of attacks\, such as “Spectre/Meltdown” and “Rowhammer”\, shattered our confidence in processors as a root of trust. Making matters worse\, modern processors are incredibly complex\, and have (as it turns out) been designed without proper attention given to security for decades. As a result\, it’s not clear the extent of the problem or where to start to fix it—forcing both attackers and defenders into an arms race whose endpoint is unclear. \nThis talk will start with an overview of my group’s work to help address the processor security problem. I will then describe a specific recent project that we published earlier this year. Finally\, I will describe an eight-year journey that led to this project\, while trying to highlight the many people responsible and the various twists and turns that our thinking took along the way.
URL:https://seasevents.nmsdev7.com/event/ese-fall-seminar-title-tbd-13/
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:20241008T101500
DTEND;TZID=America/New_York:20241008T111500
DTSTAMP:20260403T172855
CREATED:20240816T191634Z
LAST-MODIFIED:20240816T191634Z
UID:10008047-1728382500-1728386100@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Systemic Disadvantages for LGBTQ Professionals in STEM"
DESCRIPTION:Researchers have documented race and gender bias in STEM for decades\, but there has been little parallel examination of LGBTQ status as an axis of inequality. How do LGBTQ-identifying STEM professionals fare in STEM? Drawing on data from her NSF-funded STEM Inclusion Study\, which included surveys of over 25\,000 STEM workers\, Dr. Cech will discuss her research on LGBTQ inequality among STEM professionals. Her research is the first to document persistent and systemic disadvantages for LGBTQ STEM professionals\, compared to their non-LGBTQ peers. She documents these disadvantages along numerous dimensions\, including day-to-day workplace experiences\, career limitations\, professional devaluation\, marginalization and harassment\, turnover intentions\, and health and wellness issues. Her presentation will situate these findings within the broader context of intersectional inequality in STEM. This research reveals LGBTQ status as a clear axis of inequality in STEM and underscores the need for organizational and cultural shifts to address these patterns. \nThis seminar\, held in recognition of LGBT History Month and National Coming out Day\, is co-sponsored with SEAS’ Office of Diversity\, Equity and Inclusion\, the Penn LGBT Center\, Penn Out in STEM (oSTEM)\, and the Eidos LGBTQ+ Health Initiative.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-systemic-disadvantages-for-lgbtq-professionals-in-stem/
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:20241004T110000
DTEND;TZID=America/New_York:20241004T123000
DTSTAMP:20260403T172855
CREATED:20240920T201734Z
LAST-MODIFIED:20240920T201734Z
UID:10008102-1728039600-1728045000@seasevents.nmsdev7.com
SUMMARY:CBE Doctoral Dissertation Defense: "Quantitative Analysis on Enhancer-Mediated Dynamic Gene Regulations in Drosophila Embryos" (Hao Dennis Deng)
DESCRIPTION:Abstract:\nProper organism development relies on precise gene regulation\, where genes are expressed at specific times\, locations\, and quantities. Non-coding DNA sequences called enhancers play a crucial role in this process by interacting with their target promoters to initiate transcription and regulate gene expression. While the mechanisms of enhancer-mediated gene control have been extensively studied\, the dynamics of the interactions between enhancers and promoters\, especially within complex genomic environments\, remain unclear. My research employed quantitative live cell imaging in early Drosophila embryos to explore the dynamics of enhancer-promoter interactions and their impact on gene expression\, revealing several key findings. First\, we observed that strong enhancer-promoter interactions induce interallelic competition such that each homozygous allele produces fewer mRNAs than the single hemizygous allele. Through systematic gene construct modulation\, we found evidence suggesting that the transcription machinery binding to both enhancer and promoter regions\, such as RNA Pol II and pre-initiation complex molecules\, act as limiting factors that cause reduced transcription. Second\, we investigated spatiotemporal thresholds of enhancer-promoter interactions required for transcription initiation and found that enhancers need to be within 500 nm of their target promoters for more than four minutes. These spatiotemporal requirements highlight the importance of sustained proximity between enhancers and target promoters in gene activation\, providing insights into the mechanisms of precise gene control. Both studies support the hypothesis that transcription hubs form at regulatory elements\, facilitating transcription through a local concentration of transcription machinery. These findings significantly advance our understanding of the dynamics of enhancer-promoter interplay and gene regulation at the single-cell level. The knowledge can be a foundation for precise gene control in various biological contexts\, from developmental processes to potential therapeutic applications. \nZoom Link: https://upenn.zoom.us/j/94986872387?\nMeeting ID: 949 8687 2387 | Passcode: 041979
URL:https://seasevents.nmsdev7.com/event/cbe-doctoral-dissertation-defense-quantitative-analysis-on-enhancer-mediated-dynamic-gene-regulations-in-drosophila-embryos-hao-dennis-deng/
LOCATION:Room 313\, Singh Center for Nanotechnology\, 3205 Walnut 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:20241004T080000
DTEND;TZID=America/New_York:20241004T200000
DTSTAMP:20260403T172855
CREATED:20240926T071243Z
LAST-MODIFIED:20240926T071243Z
UID:10008109-1728028800-1728072000@seasevents.nmsdev7.com
SUMMARY:PRECISE Industry Day 2024 (an AI Conference)
DESCRIPTION:Theme: “Building an AI Ecosystem for a Brand New Generation of Computing”\n\n\n\n\nSection I: \n\n\nLeadership Lightning Insights Series\n\n\n\n\nSection II:\n\n\nKeynote Address\n\n\n\n\nSection III:\n\n\nExecutive Panel Discussions
URL:https://seasevents.nmsdev7.com/event/precise-industry-day-2024-an-ai-conference/
LOCATION:Wu & Chen Auditorium
ORGANIZER;CN="PRECISE":MAILTO:wng@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241002T153000
DTEND;TZID=America/New_York:20241002T163000
DTSTAMP:20260403T172855
CREATED:20240816T202240Z
LAST-MODIFIED:20240816T202240Z
UID:10008051-1727883000-1727886600@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Filling the Gaps of Machine Learning Workflows in the Molecular Sciences" (Yamil Colón\, University of Notre Dame)
DESCRIPTION:Abstract: \nOne of the main challenges in the molecular sciences is identifying promising candidates for a target application given the virtually limitless number of possibilities. Computer simulations and machine learning (ML) are integral tools to survey the vast molecular landscape and the thermodynamic conditions to evaluate their performance. Broadly\, a typical pipeline for molecular characterization\, design\, and discovery with ML involves data curation and generation\, featurization and/or descriptor development\, and a model or algorithm for the final prediction. However\, these tools are currently severely limited due to the lack of available data\, because of the time and resource expense to generate it. In this seminar\, I will discuss our research efforts focused on metal-organic frameworks (MOFs)\, which are nanoporous\, crystalline materials composed of inorganic and organic components. I will begin by showcasing our use of large-scale\, high-throughput computational screening techniques to generate gas adsorption data. Following this\, I will illustrate how transfer learning and active learning can significantly reduce the data requirements of ML models\, uncover insights into the material space\, and enhance model capabilities when combined with thermodynamic information. Finally\, I will present our recent work on utilizing sigma profiles—descriptions of molecular electrostatic surface charges—as chemical descriptors to efficiently navigate and optimize molecular landscapes.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-filling-the-gaps-of-machine-learning-workflows-in-the-molecular-sciences-yamil-colon-university-of-notre-dame/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241002T150000
DTEND;TZID=America/New_York:20241002T160000
DTSTAMP:20260403T172855
CREATED:20240919T172921Z
LAST-MODIFIED:20240919T172921Z
UID:10008101-1727881200-1727884800@seasevents.nmsdev7.com
SUMMARY:Fall 2024 GRASP SFI: Tony Samaritano\, Verge Aero\, “Building the Drone Show Industry”
DESCRIPTION:This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. \nABSTRACT\nIn this talk\, I will discuss how we engineered the foundation of the drone show industry\, creating a new form of entertainment that rivals traditional fireworks. Our journey began at Pennovation\, where we designed and tested our initial multi-drone system\, which helped us navigate the challenges of the COVID-19 pandemic. \nI will delve into the development of a scalable\, centralized control architecture for robotic agents\, enabling the safe operation of more than ten thousand drones globally. The transition from prototyping to scalable manufacturing required the tight integration of hardware and software\, including the implementation of an extensive IEEE 802.15.4 network stack on an RTOS\, and advanced flight planning software. \nSafety was paramount; sensor redundancy and reliable radio communications were crucial to creating a secure drone show system. I will also explore how balancing engineering and business imperatives—through first-principles thinking and a customer-centric approach—enabled us to navigate trade-offs among cost\, speed\, and performance effectively. \nAttendees will gain practical insights into simplifying complex systems\, making strategic trade-offs\, and employing rapid iteration to transform technical innovation into a viable market product. This talk offers valuable perspectives for engineers interested in scalable robotic systems and academics exploring the intersection of technology and entrepreneurship\, demonstrating how aligning engineering excellence with entrepreneurial vision makes success achievable.
URL:https://seasevents.nmsdev7.com/event/fall-2024-grasp-sfi-tony-samaritano/
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
END:VCALENDAR