BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Penn Engineering Events - ECPv6.15.18//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
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:20200308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20201101T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20210314T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20211107T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20220313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20221106T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210402T140000
DTEND;TZID=America/New_York:20210402T150000
DTSTAMP:20260407T155731
CREATED:20210310T131855Z
LAST-MODIFIED:20210310T131855Z
UID:10006702-1617372000-1617375600@seasevents.nmsdev7.com
SUMMARY:PICS Colloquium: "Molecular Engineering of Ice Responsive Materials: Decoding Heterogeneous Ice Nucleation"
DESCRIPTION:Abstract: The presence of particles such as dust and pollen affect cloud microphysics significantly through their effect on the state of water. These particles can hinder or accelerate the liquid-to-solid transition of water\, and also affect the ice polymorph formed in the clouds. This indirectly cloud reflectivity\, cloud lifetime\, and precipitation rates. While a predominant phenomenon\, the understanding of the surface factors that affect ice nucleation is minimal. In our research\, we use molecular simulations to illuminate the pathways through which surface properties influence ice nucleation. Experiments cannot probe the length and time scales relevant to nucleation. While molecular simulations\, in principle\, can probe the length and time scales of nucleation\, in practice nucleation is challenging to sample. Nucleation is often associated with large free energy barriers and thus\, is difficult to sample in straightforward simulations. Advanced sampling techniques and other creative approaches are needed. In this talk\, I will discuss the insights we have obtained on heterogeneous ice nucleation through studies of three surfaces – silver iodide\, kaolinite and mica. I will also highlight the synergistic combination of experiments and simulations in understanding heterogeneous ice nucleation. I will introduce a recently developed method in our group facilitate computational studies of heterogeneous nucleation. I will conclude by providing a perspective on the broader implications of our studies on interfacial phenomena and surface design.
URL:https://seasevents.nmsdev7.com/event/pics-colloquium-molecular-engineering-of-ice-responsive-materials-decoding-heterogeneous-ice-nucleation/
LOCATION:Zoom – email kathom@seas.upenn.edu
CATEGORIES:Colloquium
ORGANIZER;CN="Penn Institute for Computational Science (PICS)":MAILTO:dkparks@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210402T140000
DTEND;TZID=America/New_York:20210402T160000
DTSTAMP:20260407T155731
CREATED:20210316T190425Z
LAST-MODIFIED:20210316T190425Z
UID:10006722-1617372000-1617379200@seasevents.nmsdev7.com
SUMMARY:ESE Thesis Defense: "Accelerated Risk Assessment and Domain Adaptation for Autonomous Vehicles"
DESCRIPTION:Autonomous vehicles (AVs) are already driving on public roads around the US; however\, their rate of deployment far outpaces quality assurance and regulatory efforts. Consequently\, even the most elementary tasks\, such as automated lanekeeping\, have not been certified for safety\, and operations are constrained to narrow domains. First\, due to the limitations of worst-case analysis techniques\, we hypothesize that new methods must be developed to quantify and bound the risk of AVs. Counterintuitively\, the better the performance of the AV under consideration\, the harder it is to accurately estimate its risk as failures become rare and difficult to sample. This thesis presents a new estimation procedure and framework which can efficiently evaluate AV risk even in the rare event regime; we demonstrate the approach’s performance on a variety of AV software stacks. Second\, given a framework for AV evaluation\, we turn to a related question: how can AV software be efficiently adapted for new or expanded operating regimes? We hypothesize that stochastic search techniques can improve the naive trial-and-error approach commonly used today. One of the most challenging aspects of this task is that proficient driving requires making tradeoffs between performance and safety. Moreover\, for novel scenarios or operational domains there may be little data which can be used to understand the behavior of other drivers. To study these challenges we create a low-cost scale platform\, simulator\, benchmarks\, and baseline solutions. Using this testbed\, we develop a new population-based self play method for creating dynamic actors and detail both offline and online procedures for adapting AV components to these conditions. Taken as a whole\, this work represents a rigorous approach to the evaluation and improvement of AV software.
URL:https://seasevents.nmsdev7.com/event/ese-thesis-defense-accelerated-risk-assessment-and-domain-adaptation-for-autonomous-vehicles/
LOCATION:Zoom- Email mokelly@seas.upenn.edu for link
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:20210405T120000
DTEND;TZID=America/New_York:20210405T130000
DTSTAMP:20260407T155731
CREATED:20210120T165936Z
LAST-MODIFIED:20210120T165936Z
UID:10006590-1617624000-1617627600@seasevents.nmsdev7.com
SUMMARY:PSOC Webinar: "Cancer Cell Unjamming as Predictive Tumor Marker" (Josef A. Käs)
DESCRIPTION:Physical Sciences in Oncology Center PSOC@Penn \nSpring 2021 Webinar Series Mondays at 12:00 noon (EST) \nFor webinar links\, please contact manu@seas.upenn.edu.
URL:https://seasevents.nmsdev7.com/event/psoc-webinar-cancer-cell-unjamming-as-predictive-tumor-marker-josef-a-kas/
LOCATION:PA
CATEGORIES:Seminar
ORGANIZER;CN="PSOC":MAILTO:manu@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210406T103000
DTEND;TZID=America/New_York:20210406T120000
DTSTAMP:20260407T155731
CREATED:20210209T172446Z
LAST-MODIFIED:20210209T172446Z
UID:10006643-1617705000-1617710400@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Nonintrusive Reduced Order Models Using Physics Informed Neural Networks"
DESCRIPTION:The development of reduced order models for complex applications\, offering the promise for rapid and accurate evaluation of the output of complex models under parameterized variation\, remains a very active research area. Applications are found in problems which require many evaluations\, sampled over a potentially large parameter space\, such as in optimization\, control\, uncertainty quantification\, and in applications where a near real-time response is needed. \nHowever\, many challenges remain unresolved to secure the flexibility\, robustness\, and efficiency needed for general large-scale applications\, in particular for nonlinear and/or time-dependent problems.\nAfter giving a brief general introduction to projection based reduced order models\, we discuss the use of artificial feedforward neural networks to enable the development of fast and accurate nonintrusive models for complex problems. We demonstrate that this approach offers substantial flexibility and robustness for general nonlinear problems and enables the development of fast reduced order models for complex applications. \nIn the second part of the talk\, we discuss how to use residual based neural networks in which knowledge of the governing equations is built into the network and show that this has advantages both for training and for the overall accuracy of the model. \nTime permitting\, we finally discuss the use of reduced order models in the context of prediction\, i.e. to estimate solutions in regions of the parameter beyond that of the initial training. With an emphasis on the Mori-Zwansig formulation for time-dependent problems\, we discuss how to accurately account for the effect of the unresolved and truncated scales on the long term dynamics and show that accounting for these through a memory term significantly improves the predictive accuracy of the reduced order model. \nNB: This announcement has been updated with a new title and abstract.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-structure-preserving-reduced-order-models/
LOCATION:Zoom – Email MEAM for Link\, peterlit@seas.upenn.edu
CATEGORIES:Seminar
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210406T110000
DTEND;TZID=America/New_York:20210406T120000
DTSTAMP:20260407T155731
CREATED:20210308T181736Z
LAST-MODIFIED:20210308T181736Z
UID:10006687-1617706800-1617710400@seasevents.nmsdev7.com
SUMMARY:ESE Seminar: "Harnessing Light-Matter Interaction for Photonic Quantum Technologies"
DESCRIPTION:Photonic quantum technologies have a unique potential for applications such as large-scale quantum networks and quantum-enhanced sensing. Furthermore\, photons provide new paradigms for quantum simulations and a testbed for benchmarking the advantage of quantum simulators over the classical ones. These applications demand novel resources such as efficient single-photon sources\, large clusters of entangled photons\, and nonlinear optical gates. A 1D-atom\, a quantum emitter coupled to a single optical mode with high efficiency\, can deliver these functionalities. \nIn this talk\, I will introduce two realizations of an artificial 1D-atom based on a single quantum dot coupled to either a photonic-crystal waveguide or a tunable Fabry-Perot microcavity. To start\, I will present a single-photon source with record efficiency\, speed\, and coherence. I will then go on to discuss the transmission properties of the 1D-atom and show that we can achieve optical nonlinearities at the single-photon level. I will also show that the interplay between the local electric field and the quantum dot can lead to interesting phenomena such as directional emission and nonreciprocal transport for photons. \nAt the end of my talk\, I will present an overview of my feature research direction and my vision for quantum computing and quantum networking using photons.
URL:https://seasevents.nmsdev7.com/event/ese-seminar-harnessing-light-matter-interaction-for-photonic-quantum-technologies/
LOCATION:Zoom – Email ESE for Link jbatter@seas.upenn.edu
CATEGORIES:Seminar,Faculty,Colloquium,Student
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210406T150000
DTEND;TZID=America/New_York:20210406T160000
DTSTAMP:20260407T155731
CREATED:20210309T183147Z
LAST-MODIFIED:20210309T183147Z
UID:10006696-1617721200-1617724800@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Bridging Learning and Decision Making"
DESCRIPTION:Machine learning is becoming widely used in decision making\, in domains ranging from personalized medicine and mobile health to online education and recommendation systems. While (supervised) machine learning traditionally excels at prediction problems\, decision making requires answering questions that are counterfactual in nature\, and ignoring this mismatch leads to unreliable decisions. As a consequence\, our understanding of the algorithmic foundations for data-driven decision making is limited\, and efficient algorithms are typically developed on an ad hoc basis. Can we bridge this gap and make decision making as easy as machine learning? \nFocusing on the contextual bandit\, a core problem in data-driven decision making\, we bridge the gap by providing the first optimal and efficient reduction to supervised machine learning. The algorithm allows users to seamlessly apply off-the-shelf supervised learning models and methods to make decisions on the fly\, and has been implemented in widely-used\, industry-standard tools for decision making. \nOur results advance a broader program to develop a universal algorithm design paradigm for data-driven decision making. I will close the talk by discussing challenges and opportunities in building such a framework\, including efforts to extend our developments to difficult reinforcement learning problems in large state spaces.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-bridging-learning-and-decision-making/
LOCATION:Zoom – Email CIS for link\, cherylh@cis.upenn.edu
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210407T150000
DTEND;TZID=America/New_York:20210407T160000
DTSTAMP:20260407T155731
CREATED:20210122T023817Z
LAST-MODIFIED:20210122T023817Z
UID:10006611-1617807600-1617811200@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Engineering Microsystems and Computational Pipelines to Understand the Brain"
DESCRIPTION:Abstract \nMy lab is interested in engineering micro systems and computational tools to address questions in systems neuroscience\, developmental biology\, and cell biology that are difficult to answer with conventional techniques. We are particularly interested in the questions of how the brain is assembled during development (and changes during aging) and information processed by brain circuits. We work with a powerful genetic system – the free-living soil nematode C. elegans. In this talk\, I will introduce two sets of powerful mathematical and physics-based tools accessible by engineers to accelerate the biological understanding. I will talk about two recent developments in discrete microfluidic systems exploiting multiphase and dynamical behavior of the fluids and microswimmers (i.e. C. elegans). By designing the microfluidic system cleverly using appropriately chosen dimensionless numbers\, we can have exquisite control of the samples and experimental conditions. I will also talk about a powerful graph-theory-based framework to build probabilistic models of brain atlases. This machine-learning approach greatly reduces bias\, enables automated and robust cell identification\, and will enable a variety of applications including gene-expression analysis\, whole-brain imaging\, and connectomics.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-engineering-microsystems-and-computational-pipelines-to-understand-the-brain/
LOCATION:Zoom – Email CBE for link
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210407T150000
DTEND;TZID=America/New_York:20210407T160000
DTSTAMP:20260407T155731
CREATED:20210309T190044Z
LAST-MODIFIED:20210309T190044Z
UID:10006698-1617807600-1617811200@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Human-Centered Interactive Systems for Configuring\, Extending\, and Developing AI Applications"
DESCRIPTION:Artificial intelligence (AI) technologies are emerging and affecting our lives in many aspects. However\, the majority of individuals are merely users of AI with little capability to adapt AI to their own long-tail tasks\, preferences\, and interests that are not covered by the existing AI solutions. Democratizing AI to empower the individuals to create\, configure\, and extend AI-powered systems requires a new human-centered approach to guide the users to provide more effective inputs to the AI systems\, and to enable the systems to better understand and generalize from inputs in the user’s natural style. In this talk\, I will present my research on designing\, developing\, and studying (1) a new smartphone interactive task learning agent that empowers end-users to teach new tasks\, concepts\, and automation rules using the combination of natural language instructions and demonstrations on existing third-party app GUIs; and (2) a new developer tool that allows developers without natural language processing expertise to bootstrap a task-oriented conversational bot from existing mobile apps using the task model extracted from the app interaction traces. These projects examplify my approach of combining human-centered methods with AI techniques to create new novel interfaces and system capabilities that facilitate effective human-AI collaboration\, lowering the barriers to customizing\, extending\, and developing AI applications for end-users and non-expert developers.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-3/
LOCATION:Zoom – Email CIS for link\, cherylh@cis.upenn.edu
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210407T150000
DTEND;TZID=America/New_York:20210407T160000
DTSTAMP:20260407T155731
CREATED:20210403T155651Z
LAST-MODIFIED:20210403T155651Z
UID:10006742-1617807600-1617811200@seasevents.nmsdev7.com
SUMMARY:Spring 2021 GRASP SFI: “Robotic Caregivers—Sensing\, Simulation\, and Physical Human-Robot Interaction”
DESCRIPTION:Abstract: Autonomous robots have the potential to serve as versatile caregivers that improve quality of life for millions of people with disabilities worldwide. Yet\, physical robotic assistance presents several challenges\, including risks associated with physical human-robot interaction\, difficulty sensing the human body\, and a lack of tools for benchmarking and training physically assistive robots. In this talk\, I will present techniques towards addressing each of these core challenges in robotic caregiving. First\, I will introduce a method inspired by human perspective-taking which allow assistive robots to predict how their future actions will apply forces to a person’s body. I will then describe capacitive servoing\, a new sensing technique for robots to sense the human body and track trajectories along the body. Finally\, I will show how we can develop intelligent robotic caregivers via simulation and virtual reality\, and I will introduce Assistive Gym\, the first physics simulation framework for benchmarking and training physically assistive robots. \nClick here to join the Zoom meeting
URL:https://seasevents.nmsdev7.com/event/spring-2021-grasp-sfi-robotic-caregivers-sensing-simulation-and-physical-human-robot-interaction/
LOCATION:Zoom
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:20210408T104500
DTEND;TZID=America/New_York:20210408T114500
DTSTAMP:20260407T155731
CREATED:20210325T195944Z
LAST-MODIFIED:20210325T195944Z
UID:10006735-1617878700-1617882300@seasevents.nmsdev7.com
SUMMARY:MSE Seminar: "Bio-like Structural Hydrogels with Life-like Intelligence"
DESCRIPTION:From the cellular level up to the body system level\, living organisms present elegant designs and strategies to realize the desirable structures\, properties and functions. For example\, tendons and muscles are tough but soft\, owing to highly complex hierarchical structures rarely found in synthetic materials. Plants can automatically track the sun and our body can self-regulate motions adaptively to environment\, presenting superior intelligence also lacking in manmade systems. Hydrogels\, as a class of crosslinked polymers\, not only have tissue-like water-rich porous networks and can also change their volume and physical properties in response to environmental cues including temperature\, light\, and specific molecules. At UCLA He lab\, we exploit fundamental material processing-structure-property-function studies of hydrogels and their derivatives\, to create (i) ‘bio-like’ structures and properties and (ii) ‘life-like’ intelligence in functional soft materials for applications in robotics\, biomedicine\, energy and environment. In this talk\, I will start with showcasing how the stimuli-responsiveness of hydrogels can unify sensing-diagnosis-actuation process to create ‘synthetic intelligence’ with built-in feedback loop\, such as sunflower-like light tracking for solar harvesting (Nat. Nanotech. 2019) and self-sensing actuators for autonomous soft robotics (Sci. Robotics 2019\, 2021; Matter 2021). Driven by the remaining challenges revealed in these function developments\, I will then present our structural material approaches to breaking the fundamental limits in mechanical\, diffusion and electrical properties. I will discuss the mechanics and general principles to design extreme properties\, including simultaneously high toughness and stretchability (Nature 2021; Adv. Mater. 2021; Sci. Adv. 2020)\, tunable porosity and diffusivity (Adv. Mater. 2021; EcoMat 2021) and stretchable conductive soft materials for flexible (bio)electronics (Matter\, 2020; Adv. Funct. Mater. 2020; Adv. Mater. 2019). I will conclude my talk with a perspective on future human-machine convergence enabled by soft materials.
URL:https://seasevents.nmsdev7.com/event/mse-seminar-bio-like-structural-hydrogels-with-life-like-intelligence/
LOCATION:PA
ORGANIZER;CN="Materials Science and Engineering":MAILTO:johnruss@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210408T150000
DTEND;TZID=America/New_York:20210408T160000
DTSTAMP:20260407T155731
CREATED:20201207T171307Z
LAST-MODIFIED:20201207T171307Z
UID:10006562-1617894000-1617897600@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "Reaction-Coupled Solid-State Nanopore Digital Counting: Towards Sensitive\, Selective and Fast Nucleic Acid Testing" (Weihua Guan)
DESCRIPTION:This seminar will be held virtually on Zoom (details coming soon). \nDue to their conceptual simplicity\, the nanopore sensors have attracted intense research interest in electronic single molecule detection. While considerable success has been achieved\, the solid-state nanopores still face three significant challenges\, including repeatable nanopore size control\, introduction sensing specificity\, and prolonged sensor response time at low concentrations. In this talk\, I will discuss a calibration-free solid-state nanopore counting method and two representative applications in nucleic acid testing. One is an isothermal amplification-coupled nanopore counting for malaria analysis. The other is the CRISPR-cas12a-coupled nanopore counting for HIV analysis.  Finally\, I will also discuss how we can develop a fully integrated ‘sample-to-result’ nucleic acid testing device using the solid-state counting strategy. I believe the reaction-coupled solid-state nanopore digital counting could open a new avenue towards compact\, robust\, low-cost electronic nucleic acid testing at the point of care.
URL:https://seasevents.nmsdev7.com/event/be-seminar-weihua-guan/
LOCATION:PA
CATEGORIES:Seminar
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210408T150000
DTEND;TZID=America/New_York:20210408T160000
DTSTAMP:20260407T155731
CREATED:20210309T194731Z
LAST-MODIFIED:20210309T194731Z
UID:10006700-1617894000-1617897600@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "AI for Population Health: Melding Data and Algorithms on Networks"
DESCRIPTION:As exemplified by the COVID-19 pandemic\, our health and wellbeing depend on a difficult-to-measure web of societal factors and individual behaviors. My research aims to build AI which can impact such social challenges\, advancing health and equity on a population level. This effort requires new algorithmic and data-driven paradigms which span the full process of gathering costly data\, developing machine learning models to understand and predict interactions\, and optimizing the use of limited resources in interventions. In response to these needs\, I will present methodological developments at the intersection of machine learning\, optimization\, and social networks which are motivated by on-the-ground collaborations on HIV prevention\, tuberculosis treatment\, and the COVID-19 response. These projects have produced deployed applications and policy impact. For example\, I will present the development of an AI-augmented intervention for HIV prevention among homeless youth. This system was evaluated in a field test enrolling over 700 youth and found to significantly reduce key risk behaviors for HIV.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-ai-for-population-health-melding-data-and-algorithms-on-networks/
LOCATION:Zoom – Email CIS for link\, cherylh@cis.upenn.edu
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210409T110000
DTEND;TZID=America/New_York:20210409T123000
DTSTAMP:20260407T155731
CREATED:20210403T155854Z
LAST-MODIFIED:20210403T155854Z
UID:10006743-1617966000-1617971400@seasevents.nmsdev7.com
SUMMARY:GRASP On Robotics: “Perspectives on Machine Learning for Adaptive Robotic Systems”
DESCRIPTION:Abstract: Recent advances in machine learning are leading to new tools for designing intelligent robots: functions relied on to govern a robot’s behavior can be learned from a robot’s interaction with its environment rather than hand-designed by an engineer. Many machine learning methods assume little prior knowledge and are extremely flexible\, they can model almost anything! But\, this flexibility comes at a cost. The same algorithms are often notoriously data hungry and computationally expensive\, two problems that can be debilitating for robotics. In this talk I’ll discuss how machine learning can be combined with prior knowledge and structure to build effective solutions to robotics problems. I’ll introduce an online learning perspective on robot adaptation that unifies well-known algorithms and suggests new approaches. I’ll discuss how structure and simulation can augment learning and how imperfect models\, simple policies\, and hierarchical abstractions can help to build adaptive\, resilient systems. I will also show how we have applied some of these ideas to several robotics tasks that require impressive sensing\, speed\, and agility to complete. \nClick here to join the Zoom Webinar
URL:https://seasevents.nmsdev7.com/event/grasp-on-robotics-perspectives-on-machine-learning-for-adaptive-robotic-systems/
LOCATION:https://upenn.zoom.us/j/96715197752
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:20210409T120000
DTEND;TZID=America/New_York:20210409T130000
DTSTAMP:20260407T155731
CREATED:20210406T133212Z
LAST-MODIFIED:20210406T133212Z
UID:10006749-1617969600-1617973200@seasevents.nmsdev7.com
SUMMARY:ODEI Spotlight: Career Tools Webinar Series: Navigating the Job Search & Workplace Authentically
DESCRIPTION:Friday\, April 9th at 12pm \nRegister HERE \nOne’s identity is multi-faceted and complex\, shaped by different affinities and experiences that impact the way we see (and are seen) and operate in the world. However\, research shows that having to hide parts of your identity at work can be exhausting and mentally draining. \nThis panel will feature Penn alumni discussing strategies for both finding inclusive employers that welcome and value your identities and ways to effectively navigate the work environment as your authentic self in terms of cultural and socioeconomic backgrounds\, sexual orientation\, and lived experiences. Individuals who supervise others will also find the panel useful for tips on creating the most inclusive workplace culture possible.\nModerator: Malik Muhammad\, Associate Director\, University of Pennsylvania’s (Penn) LGBT Center. \nPanelists: \n\nMisha Chakrabarti is a speaker\, podcaster\, comedian\, and coach.\nKaren Lithgow innovates and transforms brands and businesses\nVictor Anthony Scotti\, Jr. is a DEI practitioner and Black student advocate.\n\n  \nIn partnership with Career Services and Alumni Relations
URL:https://seasevents.nmsdev7.com/event/odei-spotlight-career-tools-webinar-series-navigating-the-job-search-workplace-authentically/
LOCATION:PA
CATEGORIES:Diversity, Equity and Inclusion
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210409T140000
DTEND;TZID=America/New_York:20210409T150000
DTSTAMP:20260407T155731
CREATED:20210403T160627Z
LAST-MODIFIED:20210403T160627Z
UID:10006744-1617976800-1617980400@seasevents.nmsdev7.com
SUMMARY:Spring 2021 GRASP Seminar: “Propelling Robot Manipulation of Unknown Objects using Learned Object Centric Models”
DESCRIPTION:Abstract: There is a growing interest in using data-driven methods to scale up manipulation capabilities of robots for handling a large variety of objects. Many of these methods are oblivious to the notion of objects and they learn monolithic policies from the whole scene in image space. As a result\, they don’t generalize well to different scenes\, viewpoints\, and lighting changes. In addition\, these models cannot be combined with other components and constraints without re-training. In this talk\, I will present our approach for learning object centric models trained on 3D depth data. I will show how these approaches are combined with each other to accomplish tasks on unseen objects and environments. In particular\, I will cover our works on grasping and segmenting unknown objects\, obstacle avoidance\, and task planning for unknown object rearrangement task. \nClick here to join the Zoom meeting
URL:https://seasevents.nmsdev7.com/event/spring-2021-grasp-seminar-propelling-robot-manipulation-of-unknown-objects-using-learned-object-centric-models/
LOCATION:Zoom
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:20210409T140000
DTEND;TZID=America/New_York:20210409T160000
DTSTAMP:20260407T155731
CREATED:20210405T130937Z
LAST-MODIFIED:20210405T130937Z
UID:10006746-1617976800-1617984000@seasevents.nmsdev7.com
SUMMARY:PICS: " Houdini Presentation"
DESCRIPTION:Cinematic scientific visualization makes three dimensional scientific phenomena approachable for mass audiences by using the artistic language of film including elements like camera choreography\, lighting design\, comprehensive scenic environments\, and more. Cinematic scientific visualizations are an engaging way for domain experts to communicate niche information with the public\, to refute widely held misconceptions\, and to inspire the scientists of the future. Science films that feature these visualizations are screened at science centers to millions of viewers over the span of 10+ years and bridge different languages and cultures. They are shared widely on social media\, featured regularly in television programs\, and contribute to the success of public lectures. \nIf you are a domain expert looking to share your data more widely\, or a visualization designer who has focused on more analytical tools\, what better way is there to get started with a Hollywood style than by using Hollywood tools? This workshop will introduce participants to Houdini\, a visual effects software package that can generate cinematic-quality data visualizations with ease and efficiency. It is used and appreciated by most major animation and visual effects film studios for its procedural architecture\, its modular design\, and out-of-the-box rendering algorithms\, all important features for ease-of-use in the field of data visualization. Houdini is a general-purpose image-making software that differs from most traditional scientific visualization tools in that it is optimized for look development and design functionality. \nThis session will consist of a presentation about using scientific data in Houdini\, and a demonstration breakdown of a working Houdini scene file.\n\nNo registration is necessary.
URL:https://seasevents.nmsdev7.com/event/pics-houdini-presentation/
LOCATION:Zoom – email kathom@seas.upenn.edu
CATEGORIES:Seminar
ORGANIZER;CN="Penn Institute for Computational Science (PICS)":MAILTO:dkparks@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210412T120000
DTEND;TZID=America/New_York:20210412T130000
DTSTAMP:20260407T155731
CREATED:20210120T170209Z
LAST-MODIFIED:20210120T170209Z
UID:10006591-1618228800-1618232400@seasevents.nmsdev7.com
SUMMARY:PSOC Webinar: “Biophysical Regulation of Macrophages in Tissue Repair” (Wendy Liu)
DESCRIPTION:Physical Sciences in Oncology Center PSOC@Penn \nSpring 2021 Webinar Series Mondays at 12:00 noon (EST) \nFor webinar links\, please contact manu@seas.upenn.edu.
URL:https://seasevents.nmsdev7.com/event/psoc-webinar-wendy-liu/
LOCATION:PA
CATEGORIES:Seminar
ORGANIZER;CN="PSOC":MAILTO:manu@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210412T150000
DTEND;TZID=America/New_York:20210412T160000
DTSTAMP:20260407T155731
CREATED:20210401T181849Z
LAST-MODIFIED:20210401T181849Z
UID:10006739-1618239600-1618243200@seasevents.nmsdev7.com
SUMMARY:The Jack Keil Wolf Lecture: "Coordination of Distributed Energy Resources (DERs)"
DESCRIPTION:DERs are small-scale energy resources that can generate electricity\, such as roof-top solar PV\, store electricity\, such as Tesla’s Powerwall\, or whose electricity consumption can be flexibly controlled\, such as EV chargers\, HVACs\, water heaters\, and large industrial fans. DERs are located behind the meters of electricity consumers\, such as in homes\, commercial building and EV charging stations. With few exceptions\, DERs today are mostly uncoordinated\, that is\, they don’t operate together toward explicitly optimizing consumer or grid objectives. I will first talk about the benefits of coordinating DERs both behind-the-meter and across the meters of different electricity consumers\, and the challenges of coordinating a large number of DERs owned by different entities with different objectives and privacy constraints. I will then describe a cloud-based coordination system we developed that can overcome these challenges and its first field deployment in a California dairy farm. \nThe talk is based on joint work with Kyle Anderson\, Thomas Navidi and Professor Ram\nRajagopal’s group at Stanford.
URL:https://seasevents.nmsdev7.com/event/the-jack-keil-wolf-lecture-coordination-of-distributed-energy-resources-ders/
LOCATION:Zoom – Email ESE for Link jbatter@seas.upenn.edu
CATEGORIES:Distinguished Lecture,Faculty,Student
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210412T160000
DTEND;TZID=America/New_York:20210412T180000
DTSTAMP:20260407T155731
CREATED:20210402T134804Z
LAST-MODIFIED:20210402T134804Z
UID:10006740-1618243200-1618250400@seasevents.nmsdev7.com
SUMMARY:CIS Diversity Summit
DESCRIPTION:Help make CIS a supportive and welcoming environment. \nAll students\, faculty\, staff and postdocs are invited. \nSummit themes include increasing community in CIS; peer collaboration; how faculty can help; how to respond to microaggressions and implicit bias–both victims and bystanders; the value of a diverse TA force; productive TA-Student interaction. \nZoom link:\nhttps://upenn.zoom.us/j/95364173714?pwd=L1lnZmVXWlEyekFLSVg3OTM2bTEwdz09
URL:https://seasevents.nmsdev7.com/event/cis-diversity-summit/
LOCATION:https://upenn.zoom.us/j/96715197752
CATEGORIES:Diversity, Equity and Inclusion
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210413T100000
DTEND;TZID=America/New_York:20210413T110000
DTSTAMP:20260407T155731
CREATED:20210405T125121Z
LAST-MODIFIED:20210405T125121Z
UID:10006745-1618308000-1618311600@seasevents.nmsdev7.com
SUMMARY:ODEI Spotlight: CHOP Diversity Symposium\, Striving for Equity in STEM
DESCRIPTION:“Gaslighting in the Academy: Actually Making Black Lives Matter”\nManu Platt\, PhD\nProfessor\, Biomedical Engineering Georgia Tech and Emory University \nDate: April 13\, 2021 \nTime: 10:00 – 11:00 am \nBlueJeans Link: https://primetime.bluejeans.com/a2m/live-event/kaqbajsb \nIn honor of Diversity Month\, we are pleased to host the first annual CHOP Diversity Symposium. This is a time for us to come together and reflect on the importance of diversity\, and to highlight opportunities to make science more diverse\, equitable\, and accessible to all. This year’s theme is “Striving for Equity in STEM” and our inaugural speaker is Manu Platt\, PhD. Integrated with his research program are his mentoring goals of changing the look of the next generation of scientists and engineers to include all colors\, genders\, and backgrounds. Dr. Platt will address systemic barriers impeding underrepresented minorities in STEM\, and discuss methods to counteract the overt and implicit barriers contributing to this phenomenon.
URL:https://seasevents.nmsdev7.com/event/odei-spotlight-chop-diversity-symposium-striving-for-equity-in-stem/
LOCATION:PA
CATEGORIES:Diversity, Equity and Inclusion
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210413T103000
DTEND;TZID=America/New_York:20210413T120000
DTSTAMP:20260407T155731
CREATED:20210323T185146Z
LAST-MODIFIED:20210323T185146Z
UID:10006730-1618309800-1618315200@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Design\, Fabrication\, and Control of Biologically Inspired Soft Robots"
DESCRIPTION:Robotics has the potential to address many of today’s pressing problems in fields ranging from healthcare to manufacturing to disaster relief. However\, the traditional approaches used on the factory floor do not perform well in unstructured environments. The key to solving many of these challenges is to explore new\, non-traditional designs. Fortunately\, nature surrounds us with examples of novel ways to navigate and interact with the real world. Dr. Tolley’s Bioinspired Robotics and Design Lab seeks to borrow the key principles of operation from biological systems and apply them to robotic design. This talk will give an overview of recent projects in the lab that investigate the ways in which the use of non-rigid materials can help solve challenging problems in robotics. These projects seek to develop bioinspired systems capable of\, for example\, navigating the world by walking\, digging\, and swimming (inspired by animals like turtles\, worms\, and squid) and of interacting safely with humans and delicate objects.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-design-fabrication-and-control-of-biologically-inspired-soft-robots/
LOCATION:Zoom – Email MEAM for Link\, peterlit@seas.upenn.edu
CATEGORIES:Seminar
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210413T160000
DTEND;TZID=America/New_York:20210413T170000
DTSTAMP:20260407T155731
CREATED:20210406T182302Z
LAST-MODIFIED:20210406T182302Z
UID:10006751-1618329600-1618333200@seasevents.nmsdev7.com
SUMMARY:Neuroengineering Seminar: "Advancing Deep Brain Stimulation Therapy in Movement Disorders: from surgical implantation to behavior-based responsive therapeutic stimulation" (Enrico Opri)
DESCRIPTION:Sponsored by the Center for Neuroengineering and Therapeutics\, and the Departments of Neurosurgery and Bioengineering. \nJoin Zoom Meeting \nhttps://pennmedicine.zoom.us/j/97483734781 \nMeeting ID: 974 8373 4781 \nDeep brain stimulation (DBS) has become standard therapy for medically refractory patients with Parkinson’s disease (PD)\, essential tremor (ET)\, and other neurological disorders. The two main challenges for DBS standard-of-care are rooted in the accurate positioning of the DBS leads during intraoperative implantation and the postoperative programming of the implanted DBS device\, both needed to achieve the sought optimal therapeutic benefit. However\, both processes rely on subjective patient exams\, on expert neurophysiologists to optimize implant trajectory and programming\, and on time-consuming trial-and-error approaches. Furthermore\, existing commercially available stimulation approaches (continuous stimulation\, also known as open-loop stimulation) lack integration with patient behavior and environmental factors. We sought to address these shortcomings in the ET population\, by demonstrating the feasibility of on-demand responsive stimulation using only thalamocortical neuromarkers that modulate movement related-behavior. This led to the design and implementation of the first fully embedded closed-loop algorithm for chronic neurostimulators (CL-DBS) in humans affected by ET\, which achieved an equally effective treatment compared to current DBS approaches while having a more efficient stimulation energy profile. Furthermore\, CL-DBS demonstrated potential in decreasing DBS-related side effects (e.g. speech impairments). Additionally\, the unique window provided by intraoperative acute recordings\, allowed us to further our understanding of the thalamocortical network. We showed that there is significant cross-rhythm communication between thalamocortical regions and that changes in motor behavior correspond to changes in thalamocortical phase-amplitude coupling (PAC) profiles\, demonstrating it is a crucial mechanism for gating motor behavior. We then sought to improve intraoperative DBS implantation for the PD population by leveraging a novel biomarker\, DBS local evoked potential (DLEP)\, which strongly correlates with the location of the typical target-subregions of the nuclei of interest\, the subthalamic nucleus (STN) and globus pallidus internus (GPi). Most importantly\, the proposed methodology requires no patient interaction and could be leveraged for implementing an objective\, real-time guided placement of the DBS lead\, with a less time-consuming process and subjectivity compared to traditional mapping procedures.
URL:https://seasevents.nmsdev7.com/event/neuroengineering-seminar-advancing-deep-brain-stimulation-therapy-in-movement-disorders-from-surgical-implantation-to-behavior-based-responsive-therapeutic-stimulation-enrico-opri/
LOCATION:https://upenn.zoom.us/j/96715197752
CATEGORIES:Seminar,Postdoctoral
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210414T150000
DTEND;TZID=America/New_York:20210414T160000
DTSTAMP:20260407T155731
CREATED:20210122T024712Z
LAST-MODIFIED:20210122T024712Z
UID:10006612-1618412400-1618416000@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Robots That Eat\, Breathe and Bleed"
DESCRIPTION:Abstract \nModern robots lack the multifunctional\, interconnected chemical systems found in living organisms and\, consequently\, exhibit reduced efficiency and autonomy. At the same time\, new advancements in chemistry are enabling synthetic materials with capabilities that surpass biological materials. This talk will discuss how advances in electrochemistry and soft materials can transform the way we build and use robots\, with the ultimate goal of surpassing the capabilities of living organisms. Specifically\, fundamental insights will be applied to improve the performance and capabilities of structural materials and energy storage in robots. The discussion on energy storage will cover materials and manufacturing techniques\, funded by DARPA’s SHRIMP program\, that triple the energy density of batteries for small scale robots\, approaches that break the scaling laws of energy storage technologies by allowing robots to eat metal in their environment\, and multifunctional synthetic vascular systems that increase the energy density of robots by up to 4x that of robots that only use lithium ion batteries. The talk will finish with a discussion on how old tools like self-assembly enable metals with the strength of titanium and density of water\, and how new approaches to healing\, through transport mediated in water\, allow bone-inspired room-temperature healing of metals.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-robots-that-eat-breathe-and-bleed/
LOCATION:Zoom – Email CBE for link
CATEGORIES:Seminar
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210414T150000
DTEND;TZID=America/New_York:20210414T160000
DTSTAMP:20260407T155731
CREATED:20210318T160456Z
LAST-MODIFIED:20210318T160456Z
UID:10006727-1618412400-1618416000@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Digital humans that move\, interact\, express\, and feel\, just like us!"
DESCRIPTION:The creation of intelligent virtual agents (IVAs) or digital humans is vital for many virtual and augmented reality systems. As the world increasingly uses digital and virtual platforms for everyday communication and interactions\, there is a heightened need to create human-like virtual avatars and agents endowed with social and emotional intelligence. Interactions between humans and virtual agents are being used in different areas including\, VR\, games and story-telling\, computer-aided design\, social robotics\, and healthcare. Designing and building intelligent agents that can communicate and connect with people is necessary but not sufficient. Researchers must also consider how these IVAs will inspire trust and desire among humans. Knowing the perceived affective states and social-psychological constructs (such as behavior\, emotions\, psychology\, motivations\, and beliefs) of humans in such scenarios allows the agents to make more informed decisions and navigate and interact in a socially-intelligent manner. \nWhen humans communicate with each other\, they frequently use a combination of verbal communications\, interpersonal relationships between the people involved\, and non-verbal body expressions during communication to complement\, contradict\, substitute\, or regulate what is being said. These signals are essential for understanding each other\, particularly regarding expressing changing moods and affective states. In this talk\, I will give an overview of our recent work on simulating intelligent\, interactive\, and immersive human-like agents who can also learn\, understand and be sentient to the world around them using a combination of emotive gestures\, gaits\, and expressions. This problem arises in the context of human motion and behavior computation that takes environmental\, sociological\, and psychological factors into account. Finally\, I will also talk about our many on-going projects which use our AI-driven IVAs\, including intelligent digital humans for urban simulation\, crowd simulation\, mental health and therapy applications\, and social robotics
URL:https://seasevents.nmsdev7.com/event/cis-seminar-digital-humans-that-move-interact-express-and-feel-just-like-us/
LOCATION:Zoom – Email CIS for link\, cherylh@cis.upenn.edu
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210414T173000
DTEND;TZID=America/New_York:20210414T190000
DTSTAMP:20260407T155731
CREATED:20210406T134113Z
LAST-MODIFIED:20210406T134113Z
UID:10006750-1618421400-1618426800@seasevents.nmsdev7.com
SUMMARY:ODEI Spotlight: Race\, Solidarity\, and Allyship
DESCRIPTION:Moderated by:\nMalik Muhammad\nAssoc. Director – LGBT Center \nPanelists:\nDr. Valerie Dorsey Allen\nDirector – African American Resource Center \nElizabeth Cannon\nCo-Founder – White Educators Committed to Anti-Racism & Equity \n  \nZoom ID: 970 2429 7938
URL:https://seasevents.nmsdev7.com/event/odei-spotlight-race-solidarity-and-allyship/
LOCATION:PA
CATEGORIES:Diversity, Equity and Inclusion
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210415T100000
DTEND;TZID=America/New_York:20210415T113000
DTSTAMP:20260407T155731
CREATED:20210414T170809Z
LAST-MODIFIED:20210414T170809Z
UID:10006763-1618480800-1618486200@seasevents.nmsdev7.com
SUMMARY:CBE PhD Dissertation Defense | A Computational Approach to the Study of Trauma
DESCRIPTION:Abstract: \nTrauma with hypovolemic shock is an extreme pathological state that challenges the body to maintain blood pressure and oxygenation in the face of hemorrhagic blood loss. In conjunction with surgical actions and transfusion therapy\, survival requires the patient’s blood to maintain hemostasis to stop bleeding. The physics of the problem are multiscale: (1) the systemic circulation sets the global blood pressure in response to blood loss and resuscitation therapy\, (2) local tissue perfusion is altered by localized vasoregulatory mechanisms and bleeding\, and (3) altered blood and vessel biology resulting from the trauma as well as local hemodynamics control the assembly of clotting components at the site of injury. Building upon ongoing modeling efforts to simulate arterial or venous thrombosis in a diseased vasculature\, we have developed models of trauma (both multiscale and machine-learning based) to understand patient risk and predict response. Key results were: (1) the upstream vascular network rapidly depressurizes to reduce blood loss\, (2) wall shear rates at the hemorrhaging wound exit are sufficiently high (~10\,000 s-1) to drive von Willebrand Factor unfolding\, (3) full coagulopathy results in >2L blood loss in 2 hours for severing all vessels of 0.13 to 0.005 mm diameter within the bifurcating network\, whereas full hemostasis limits blood loss to <100 mL within 2 min\, and (4) hemodilution from transcapillary refill increases blood loss and could be implicated in trauma induced coagulopathy. Machine learning based methods were also implemented to understand trauma patient outcomes. A 400-estimator gradient boosting classifier was trained to predict survival probability and the model is able to predict a survival probability for any trauma patient and accurately distinguish between a deceased and survived patient in 92.4% of all cases. Partial dependence curves (Psurvival vs. feature value) obtained from the trained model revealed the global importance of Glasgow coma score\, age\, and systolic blood pressure while pulse rate\, respiratory rate\, temperature\, oxygen saturation\, and gender had more subtle single variable influences. Shapley values\, which measure the relative contribution of each of the 8 features to individual patient risk\, were computed for several patients and quantified patient-specific warning signs.
URL:https://seasevents.nmsdev7.com/event/cbe-phd-dissertation-defense-a-computational-approach-to-the-study-of-trauma/
LOCATION:Zoom – Email CBE for link
CATEGORIES:Doctoral,Graduate,Dissertation or Thesis Defense
ORGANIZER;CN="Chemical and Biomolecular Engineering":MAILTO:cbemail@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210415T104500
DTEND;TZID=America/New_York:20210415T114500
DTSTAMP:20260407T155731
CREATED:20210407T221212Z
LAST-MODIFIED:20210407T221212Z
UID:10006752-1618483500-1618487100@seasevents.nmsdev7.com
SUMMARY:MSE Doctoral Student Seminar
DESCRIPTION:
URL:https://seasevents.nmsdev7.com/event/mse-doctoral-student-seminar/
LOCATION:PA
ORGANIZER;CN="Materials Science and Engineering":MAILTO:johnruss@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210415T150000
DTEND;TZID=America/New_York:20210415T160000
DTSTAMP:20260407T155731
CREATED:20201207T171501Z
LAST-MODIFIED:20201207T171501Z
UID:10006563-1618498800-1618502400@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "Understanding Spatiotemporal Cell Reprogramming for Precision Medicine" (Xiling Shen)
DESCRIPTION:This seminar will be held virtually on zoom (check email for link or contact ksas@seas.upenn.edu). \nBodily cells undergo transformations in space and time during development\, disease progression\, and therapeutic treatment. A holistic approach that combines engineering tools\, patient-derived models\, and analytical methods is needed to map cellular reprogramming and expose new therapeutic opportunities. The talk will cover our effort across the entire spectrum from bench to bedside\, including organogenesis during embryonic development\, epigenetic and metabolic reprogramming of cancer metastasis and COVID-19 patients\, and organoid technology to guide precision- and immune-oncology.
URL:https://seasevents.nmsdev7.com/event/be-seminar-xiling-shen/
LOCATION:PA
CATEGORIES:Seminar
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210415T150000
DTEND;TZID=America/New_York:20210415T160000
DTSTAMP:20260407T155731
CREATED:20210309T195525Z
LAST-MODIFIED:20210309T195525Z
UID:10006701-1618498800-1618502400@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Exploiting latent structure and bisimulation metrics for better generalization in reinforcement learning"
DESCRIPTION:The advent of deep learning has shepherded unprecedented progress in various fields of machine learning. Despite recent advances in deep reinforcement learning (RL) algorithms\, however\, there is no method today that exhibits anywhere near the generalization that we have seen in computer vision and NLP. Indeed\, one might ask whether deep RL algorithms are even capable of the kind of generalization that is needed for open-world environments.  This challenge is fundamental and will not be solved with incremental algorithmic advances.  \nIn this talk\, we propose to incorporate different assumptions that better reflect the real world and allow the design of novel algorithms with theoretical guarantees to address this fundamental problem. We first present how state abstractions can accelerate reinforcement learning from rich observations\, such as images\, without relying either on domain knowledge or pixel-reconstruction. Our goal is to learn state abstractions that both provide for effective downstream control and invariance to task-irrelevant details. We use bisimulation metrics to quantify behavioral similarity between states\, and learn robust latent representations which encode only the task-relevant information from observations. We provide theoretical guarantees for the learned approximate abstraction and extend this notion to families of tasks with varying dynamics.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-exploiting-latent-structure-and-bisimulation-metrics-for-better-generalization-in-reinforcement-learning/
LOCATION:Zoom – Email CIS for link\, cherylh@cis.upenn.edu
ORGANIZER;CN="Computer and Information Science":MAILTO:cherylh@cis.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20210416T110000
DTEND;TZID=America/New_York:20210416T123000
DTSTAMP:20260407T155731
CREATED:20210411T205022Z
LAST-MODIFIED:20210411T205022Z
UID:10006758-1618570800-1618576200@seasevents.nmsdev7.com
SUMMARY:GRASP On Robotics: “Learning and Games in Self-Driving”
DESCRIPTION:Abstract: In this talk\, I’ll explore the power of a game-theoretic viewpoint in self-driving and in machine learning. We begin by considering the application of machine learning to Aurora’s advanced self-driving system in both perception and decision making. We discuss complexities that arise from multi-actor interaction. \nWe then explore the\, perhaps surprising\, role a game-theoretic view can take in developing algorithms for learning to make decisions. In particular\, we review a “no-regret” game-theoretic perspective on model-based RL\, Approximate Policy Iteration\, and Inverse Optimal Control. \nClick here to join the Zoom Webinar
URL:https://seasevents.nmsdev7.com/event/grasp-on-robotics-learning-and-games-in-self-driving/
LOCATION:PA
END:VEVENT
END:VCALENDAR