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: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:20240201T153000
DTEND;TZID=America/New_York:20240201T163000
DTSTAMP:20260403T173600
CREATED:20240108T171010Z
LAST-MODIFIED:20240108T171010Z
UID:10007792-1706801400-1706805000@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "Imaging the Brain for Deeper\, Finer\, and More Diverse Insight" (Fei Xia\, Ecole Normale Supérieure)
DESCRIPTION:The brain is a powerful organ that regulates health and drives intelligence. How can we see more clearly into the working brain to understand it better? High-resolution optical microscopy has long been an ideal tool due to its spatial sub-micron precision and specificity. However\, achieving the necessary spatio-temporal scale for further advancing our understanding of the brain remains a challenge. The primary obstacle originates from the inhomogeneous nature of tissues\, which distort light and compromise imaging depth\, precision\, and information. \nIn this talk\, I will introduce new in vivo deep tissue optical microscopy platforms advanced in microscope design and computational tools\, to address existing challenges. By moving towards longer wavelengths for one-\, two-\, and three-photon microscopy\, we have achieved simultaneous deep structural and functional imaging through an entire cortical column with multiple label-free and fluorescence contrasts. With precise control of light\, we have enabled adaptive optical imaging of dendritic spines and myelinated axons up to the hippocampus. With a new AI-enabled tool\, we have sped up volumetric 3D imaging for microvasculature\, neurons\, and dendrites. These techniques advance toward imaging the dynamic\, cell-type-specific processes and microvasculature within the living brain. I will conclude by discussing the opportunities these enabling optical microscopy techniques offer for biology and clinical applications.
URL:https://seasevents.nmsdev7.com/event/be-seminar-fei-xia-ecole-normale-superieure/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240202T103000
DTEND;TZID=America/New_York:20240202T114500
DTSTAMP:20260403T173600
CREATED:20240126T155547Z
LAST-MODIFIED:20240126T155547Z
UID:10007831-1706869800-1706874300@seasevents.nmsdev7.com
SUMMARY:Spring 2024 GRASP on Robotics: John Doyle\, California Institute of Technology\, "Control/Architecture/Control"
DESCRIPTION:This is a hybrid event with a VIRTUAL SPEAKER. There will be in-person attendance in Wu and Chen and virtual attendance on Zoom. \nABSTRACT\nThis talk will describe progress in developing a universal theory of architectures for complex networks\, motivated by and applied to aerospace\, process control\, internet\, cyberphysical\, ecosystems\, multiscale physics\, turbulence\, biology\, neuroscience\, medicine\, linguistics\, and social systems.
URL:https://seasevents.nmsdev7.com/event/spring-2024-grasp-on-robotics-john-doyle-california-institute-of-technology-control-architecture-control/
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:20240202T140000
DTEND;TZID=America/New_York:20240202T150000
DTSTAMP:20260403T173600
CREATED:20240123T174255Z
LAST-MODIFIED:20240123T174255Z
UID:10007821-1706882400-1706886000@seasevents.nmsdev7.com
SUMMARY:PICS Colloquium: "Exploring the landscape of model representations"
DESCRIPTION:Many studies adopt low-resolution\, coarse-grained (CG) models to investigate polymers\, proteins\, and other soft materials. These studies must first specify the details that are retained in the low-resolution model\, i.e.\, they must specify the “CG representation.” Unfortunately\, the “best” representation for complex systems is not always obvious. In this study\, we systematically explore the space of model representations for a typical protein and we examine how the properties of the CG model depend upon the choice of representation\, i.e.\, the details retained in the CG model. By adopting a simple high-resolution model for protein fluctuations\, we quantitatively assess the quality of a representation based upon its information content\, I\, and spectral quality\, Q. While I quantifies the information lost due to eliminating details from the high-resolution model\, Q quantifies the extent to which the representation preserves large scale motions. By employing these metrics as energy functions and adopting an ergodic move set\, we explore the local and global minima in the space of representations. Additionally\, by employing Monte Carlo methods\, we quantify the number of representations with a given quality. We find that representations with high spectral quality match our physical intuition\, while highly informative representations do not. Indeed\, we find that the information content and spectral quality are anti-correlated among low-resolution representations. Moreover\, our study suggests the possibility of a critical resolution below which there may exist a “phase transition” distinguishing good and bad representations. These studies may provide insight for developing CG models of soft materials and\, more generally\, for developing reduced representations of complex phenomena or high-dimensional data.
URL:https://seasevents.nmsdev7.com/event/pics-colloquium-exploring-the-landscape-of-model-representations/
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:20240205T153000
DTEND;TZID=America/New_York:20240205T163000
DTSTAMP:20260403T173600
CREATED:20231219T204537Z
LAST-MODIFIED:20231219T204537Z
UID:10007786-1707147000-1707150600@seasevents.nmsdev7.com
SUMMARY:Penn Engineering 2023-24 Pender Award Lecture: Shafi Goldwasser
DESCRIPTION:
URL:https://seasevents.nmsdev7.com/event/penn-engineering-2023-24-pender-award-lecture-shafi-goldwasser/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar,Distinguished Lecture,Faculty
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240206T100000
DTEND;TZID=America/New_York:20240206T113000
DTSTAMP:20260403T173600
CREATED:20240116T230557Z
LAST-MODIFIED:20240116T230557Z
UID:10007816-1707213600-1707219000@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: “Towards the Discovery of Trustworthy and Interpretable ML-enabled Constitutive Laws for Solids from Low- and Limited-data”
DESCRIPTION:Machine learning techniques are gearing up to play a significant role in the field of computational solid mechanics and multiphysics\, enabling the integration of experimental data and physical constraints towards data-driven constitutive laws\, acceleration of computational techniques for multi-scale modeling\, and new paradigms for the solution of forward and inverse problems\, to name a few. This talk will cover recent advancements in the area of ML-enabled constitutive modeling I) A physics-informed data-driven constitutive modeling approach for isotropic and anisotropic hyperelastic materials is developed using tensor representation theorems. The trained surrogates (using GPR and NNs) are able to respect physical principles such as material frame indifference\, material symmetry\, and the local balance of angular momentum. Overall\, the presented approach is tested on synthetic data from isotropic and anisotropic constitutive laws and shows surprising accuracy even far beyond the limits of the training domain\, indicating that the resulting surrogates can efficiently generalize as they incorporate knowledge about the underlying physics. Additionally it is shown that the inherent material symmetries can be discovered directly from data. II) Finally\, we proceed to tackle elastoplasticity in a modular framework. Employing convexity for the yield functions we recover textrure-parametrized yield functions using input convex (IC)NNs and propose a hybrid model-data-driven framework to recover yield functions with tension compression asymmetries in the low data regime. Additionally by employing thermodynamic requirements in an NN-based framework we learn hardening laws from limited experimental data. III) The extension of the above approaches for both hyperelasticity and elastoplasticity to enable interpretable discovery of constitutive models without the use of model libraries will be discussed. IV) Finally ML-hybrid approaches for the solution of PDEs will be introduced\, focusing on applications in solid and structural mechanics towards high dimensional optimization problems requiring robust and trustworthy solvers.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-towards-the-discovery-of-trustworthy-and-interpretable-ml-enabled-constitutive-laws-for-solids-from-low-and-limited-data/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240206T153000
DTEND;TZID=America/New_York:20240206T163000
DTSTAMP:20260403T173600
CREATED:20240124T182229Z
LAST-MODIFIED:20240124T182229Z
UID:10007824-1707233400-1707237000@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: " How Algorithms Can Support Deliberative Democracy"
DESCRIPTION:Academics and political practitioners around the world are experimenting with a class of democratic innovations called deliberative mini-publics (DMs). In a DM\, a panel of constituents convenes to deliberate about specific issues and make policy recommendations to traditional political decision-makers (e.g.\, legislators). Nearly all DMs rely on sortition – random selection – to choose the panelists. Sortition is often thought of as a simple lottery that chooses all constituents with equal probability. In practice\, however\, simple random selection fails to yield representative panels due to selection bias in who accepts invitations to participate. Many practitioners of DMs therefore sacrifice the pure equality embodied by a simple lottery\, instead imposing quotas on socially salient groups and then “randomizing” within those constraints. \nEngineering this randomization within user-specified quotas turns out to be technically demanding. My talk covers our algorithmic solution to this problem: a framework of optimization-based algorithms which\, subject to such quotas\, ensure individuals’ selection probabilities are as equal as possible\, as measured by any convex function measuring equality (Fair Algorithms for Selecting Citizens’ Assemblies\, Nature\, ‘21). \nAfter presenting our approach to this technical problem\, I discuss my follow-up work demonstrating how the notion of equality we choose to optimize within this framework has implications for normative goals like fairness\, transparency\, and resistance to subversion. This includes a discussion of Leximin\, the original instantiation of our framework\, which has been adopted widely in practice and is available for public use at Panelot.org.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-how-algorithms-can-support-deliberative-democracy/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240207T103000
DTEND;TZID=America/New_York:20240207T113000
DTSTAMP:20260403T173600
CREATED:20240202T144841Z
LAST-MODIFIED:20240202T144841Z
UID:10007846-1707301800-1707305400@seasevents.nmsdev7.com
SUMMARY:Spring 2024 GRASP SFI: Andrew Owens\, University of Michigan\, "Multimodal Learning from the Bottom Up"
DESCRIPTION:This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. \nABSTRACT\nToday’s machine perception systems rely extensively on supervision provided by humans\, such as natural language. I will talk about our efforts to make systems that\, instead\, learn from two ubiquitous sources of unlabeled sensory data: visual motion and cross-modal associations between senses. First\, I will discuss our work on creating unified self-supervised motion analysis methods that can address both object tracking and optical flow tasks. I will then discuss how these same techniques can be applied to localizing sound sources in video\, and how tactile sensing data can be used to train multimodal  visual-tactile models. Finally\, I will talk about our recent work on subverting visual perception systems\, by creating “multi-view” optical illusions: images that change their appearance under a transformation\, such as a flip or rotation.
URL:https://seasevents.nmsdev7.com/event/spring-2024-grasp-sfi-andrew-owens/
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:20240207T120000
DTEND;TZID=America/New_York:20240207T133000
DTSTAMP:20260403T173600
CREATED:20240125T211658Z
LAST-MODIFIED:20240125T211658Z
UID:10007828-1707307200-1707312600@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "Paths to AI Accountability" (Sarah Cen\, Massachusetts Institute of Technology)
DESCRIPTION:ABSTRACT: \nIn the past decade\, we have begun grappling with difficult questions related to the rise of AI\, including: What rights do individuals have in the age of AI? When should we regulate AI and when should we abstain? What degree of transparency is needed to monitor AI systems? These questions are all concerned with AI accountability: determining who owes responsibility and to whom in the age of AI. In this talk\, I will discuss the two main components of AI accountability\, then illustrate them through a case study on social media. Within the context of social media\, I will focus on how social media platforms filter (or curate) the content that users see. I will review several methods for auditing social media\, drawing from concepts and tools in hypothesis testing\, causal inference\, and LLMs. \n  \nZOOM LINK (if unable to attend in-person): https://upenn.zoom.us/j/96315573573 \n 
URL:https://seasevents.nmsdev7.com/event/asset-seminar-sarah-cen-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:20240207T153000
DTEND;TZID=America/New_York:20240207T163000
DTSTAMP:20260403T173600
CREATED:20240116T175613Z
LAST-MODIFIED:20240116T175613Z
UID:10007806-1707319800-1707323400@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Molecular Microscopy with Single Cell Transcriptomic Data Resolves RNA Liquid Biopsies" (Sevahn Vorperian\, Stanford)
DESCRIPTION:Abstract\nInvasive biopsy is the gold standard for diagnosing several diseases; however\, these procedures offer a limited\, localized view of the disease pathology to the physician and are not risk-free to the patient. Cell-free RNA (cfRNA) in blood plasma reflects dynamic gene expression changes and can facilitate early disease diagnosis\, yet current cfRNA assays fall short of the cellular resolution afforded by an invasive biopsy. In this talk\, I will first resolve plasma cfRNA at cell type resolution using single cell transcriptomic data alongside approaches from machine learning and data science\, which enable a systems-view into the underlying molecular patterns within these high-dimensional biological data. I will then describe how this molecular microscope can noninvasively reflect changes observed in invasive biopsy across various diseases and facilitate the study of biofluids beyond the blood. These findings expand the achievable resolution within the RNA liquid biopsy biomolecular repertoire and broaden opportunities in precision medicine for complex diseases in hard-to-biopsy tissues.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-molecular-microscopy-with-single-cell-transcriptomic-data-resolves-rna-liquid-biopsies-sevahn-vorperian-stanford/
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:20240208T150000
DTEND;TZID=America/New_York:20240208T170000
DTSTAMP:20260403T173600
CREATED:20240201T155129Z
LAST-MODIFIED:20240201T155129Z
UID:10007845-1707404400-1707411600@seasevents.nmsdev7.com
SUMMARY:Evolution of Data Storytelling: Women in Data Science x Penn Museum Tour + Workshop
DESCRIPTION:Join us for an exciting kick-off event at the Penn Museum as part of the Women in Data Science (WiDS) @ Penn conference\, where the past meets the future in a guided tour and storytelling workshop.
URL:https://seasevents.nmsdev7.com/event/evolution-of-data-storytelling-women-in-data-science-x-penn-museum-tour-workshop/
LOCATION:The Penn Museum\, 3260 South St\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Workshop
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240208T153000
DTEND;TZID=America/New_York:20240208T163000
DTSTAMP:20260403T173600
CREATED:20240116T154201Z
LAST-MODIFIED:20240116T154201Z
UID:10007803-1707406200-1707409800@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "Mapping and engineering gene expression with chemical and spatial lenses" (Hailing Shi\, Broad Institute & MIT)
DESCRIPTION:Precise RNA expression\, tailored to specific brain regions\, cell types\, and subcellular compartments\, is pivotal for orchestrating complex brain functions. In the first part of my talk\, I will introduce a confocal imaging-based spatial transcriptomics platform\, STARmap\, that seamlessly combines in situ hybridization\, hydrogel tissue chemistry\, and in situ sequencing technologies. Leveraging scalable experimental and computational pipelines\, we have constructed a comprehensive spatial cell atlas of the mouse brain\, revealing subregion-specific cell types\, previously undiscovered tissue architectures\, and viral tropisms. The second part of my talk transits to chemical modifications on messenger RNAs (mRNAs) and their far-reaching implications in gene expression regulation and the development of RNA therapeutics. I will discuss the messenger-oligonucleotide conjugate RNAs (mocRNAs) design\, showcasing how we can harness chemical modifications for engineering enhanced gene delivery vectors. Looking into the future\, I aim to innovate and integrate chemical and spatial profiling approaches to understand tissue health and disease in-depth.
URL:https://seasevents.nmsdev7.com/event/be-seminar-mapping-and-engineering-gene-expression-with-chemical-and-spatial-lenses-hailing-shi-broad-institute-mit/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Seminar
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240208T153000
DTEND;TZID=America/New_York:20240208T163000
DTSTAMP:20260403T173600
CREATED:20240129T163131Z
LAST-MODIFIED:20240129T163131Z
UID:10007832-1707406200-1707409800@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Rethinking Data Use in Large Language Models"
DESCRIPTION:Large language models (LMs) such as ChatGPT have revolutionized natural language processing and artificial intelligence more broadly. In this talk\, I will discuss my research on understanding and advancing these models\, centered around how they use the very large text corpora they are trained on. First\, I will describe our efforts to understand how these models learn to perform new tasks after training\, demonstrating that their so-called in context learning capabilities are almost entirely determined by what they learn from the training data. Next\, I will introduce a new class of LMs—nonparametric LMs—that repurpose this training data as a data store from which they retrieve information for improved accuracy and updatability. I will describe my work on establishing the foundations of such models\, including one of the first broadly used neural retrieval models and an approach that simplifies a traditional\, two-stage pipeline into one. I will also discuss how nonparametric models open up new avenues for responsible data use\, e.g.\, by segregating permissive and copyrighted text and using them differently. Finally\, I will envision the next generation of LMs we should build\, focusing on efficient scaling\, improving factuality\, and decentralization.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-rethinking-data-use-in-large-language-models/
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:20240209T083000
DTEND;TZID=America/New_York:20240209T150000
DTSTAMP:20260403T173600
CREATED:20240201T155601Z
LAST-MODIFIED:20240201T155601Z
UID:10007844-1707467400-1707490800@seasevents.nmsdev7.com
SUMMARY:Women in Data Science @ Penn
DESCRIPTION:The Wharton School and Penn Engineering are proud to host the fifth annual Women in Data Science (WiDS) @ Penn Conference on February 8-9\, 2024\, on the University of Pennsylvania’s campus. A celebrated interdisciplinary event\, WiDS @ Penn welcomes academic\, industry\, and student speakers from across the data science landscape to celebrate its diversity\, both in subject matter and personnel.
URL:https://seasevents.nmsdev7.com/event/women-in-data-science-penn/
LOCATION:Jon M. Huntsman Hall\, 3730 Walnut Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Conference
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240209T103000
DTEND;TZID=America/New_York:20240209T114500
DTSTAMP:20260403T173600
CREATED:20231120T162119Z
LAST-MODIFIED:20231120T162119Z
UID:10007762-1707474600-1707479100@seasevents.nmsdev7.com
SUMMARY:Spring 2024 GRASP on Robotics: Qixing Huang\, University of Texas at Austin\, "Geometric Regularizations for 3D Shape Generation"
DESCRIPTION:This is a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. \nABSTRACT\nGenerative models\, which map a latent parameter space to instances in an ambient space\, enjoy various applications in 3D Vision and related domains. A standard scheme of these models is probabilistic\, which aligns the induced ambient distribution of a generative model from a prior distribution of the latent space with the empirical ambient distribution of training instances. While this paradigm has proven to be quite successful on images\, its current applications in 3D generation encounter fundamental challenges in the limited training data and generalization behavior. The key difference between image generation and shape generation is that 3D shapes possess various priors in geometry\, topology\, and physical properties. Existing probabilistic 3D generative approaches do not preserve these desired properties\, resulting in synthesized shapes with various types of distortions. In this talk\, I will discuss recent work that seeks to establish a novel geometric framework for learning shape generators. The key idea is to model various geometric\, physical\, and topological priors of 3D shapes as suitable regularization losses by developing computational tools in differential geometry and computational topology. We will discuss the applications in deformable shape generation\, latent space design\, joint shape matching\, and 3D man-made shape generation.
URL:https://seasevents.nmsdev7.com/event/spring-2024-grasp-on-robotics-qixing-huang-university-of-texas-at-austin-geometric-regularizations-for-3d-shape-generation/
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:20240209T140000
DTEND;TZID=America/New_York:20240209T150000
DTSTAMP:20260403T173600
CREATED:20240130T141449Z
LAST-MODIFIED:20240130T141449Z
UID:10007835-1707487200-1707490800@seasevents.nmsdev7.com
SUMMARY:PICS Colloquium: “Wall-models of turbulent flows via scientific multi-agent reinforcement learning”
DESCRIPTION:The predictive capabilities of turbulent flow simulations\, critical for aerodynamic design and weather prediction\, hinge on the choice of turbulence models. The abundance of data from experiments and simulations and the advent of machine learning have provided a boost to turbulence modeling efforts. However\, simulations of turbulent flows remain hindered by the inability of heuristics and supervised learning to model the near-wall dynamics. We address this challenge by introducing scientific multi-agent reinforcement learning (SciMARL) for the discovery of wall models for large-eddy simulations (LES). In SciMARL\, discretization points act also as cooperating agents that learn to supply the LES closure model. The agents self-learn using limited data and generalize to higher Reynolds numbers in reproducing key flow quantities. We test the discovered wall model to canonical flat plate boundary layers\, which shows good predictable capabilities outside the Reynolds numbers used to train the model. We will discuss extensions to this model for flows with pressure-gradient effects.
URL:https://seasevents.nmsdev7.com/event/pics-colloquium-wall-models-of-turbulent-flows-via-scientific-multi-agent-reinforcement-learning/
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:20240212T110000
DTEND;TZID=America/New_York:20240212T120000
DTSTAMP:20260403T173600
CREATED:20240201T134202Z
LAST-MODIFIED:20240201T134202Z
UID:10007836-1707735600-1707739200@seasevents.nmsdev7.com
SUMMARY:ESE Spring Seminar - "Physics-inspired Machine Learning"
DESCRIPTION:Combining physics with machine learning is a rapidly growing field of research. Thereby\, most work focuses on leveraging machine learning methods to solve problems in physics. Here\, however\, we focus on the converse\, i.e.\, physics-inspired machine learning\, which can be described as incorporating structure from physical systems into machine learning methods to obtain models with better inductive biases. More concretely\, we propose several physics-inspired deep learning architectures for sequence modelling based on nonlinear coupled oscillators\, Hamiltonian systems and multi-scale dynamical systems. The proposed architectures tackle central problems in the field of recurrent sequence modeling\, namely the vanishing and exploding gradients problem as well as the issue of insufficient expressive power. Moreover\, we discuss physics-inspired learning on graphs\, wherein the dynamics of the message-passing propagation are derived from physical systems. We further prove that these methods mitigate the over-smoothing issue\, thereby enabling the construction of deep graph neural networks (GNNs). We extensively test all proposed methods on a variety of versatile synthetic and real-world datasets\, ranging from image recognition\, speech recognition\, natural language processing (NLP)\, medical applications\, and scientific computing for sequence models\, to citation networks\, computational chemistry applications\, and networks of articles and websites for graph learning models. Finally\, we show how to leverage physics-based inductive biases of physics-inspired machine learning methods to solve problems in the physical sciences.
URL:https://seasevents.nmsdev7.com/event/ese-spring-seminar-tbd/
LOCATION:Glandt Forum\, Singh Center for Nanotechnology\, 3205 Walnut 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:20240212T130000
DTEND;TZID=America/New_York:20240212T150000
DTSTAMP:20260403T173600
CREATED:20240112T192252Z
LAST-MODIFIED:20240112T192252Z
UID:10007801-1707742800-1707750000@seasevents.nmsdev7.com
SUMMARY:BE Doctoral Dissertation Defense: "Computational imaging and multiomic biomarkers for precision medicine: characterizing heterogeneity in lung cancer" (Apurva Singh)
DESCRIPTION:The Department of Bioengineering at the University of Pennsylvania and Dr. Despina Kontos are pleased to announce the Doctoral Dissertation Defense of Apurva Singh.\n\nTitle:  “Computational imaging and multiomic biomarkers for precision medicine: characterizing heterogeneity in lung cancer”\nDate: February 12\, 2024\nTime: 1:00 PM\nLocation:  John Morgan Building\, Class of 62 auditorium\n\nZoom link\n\nThe public is welcome to attend.
URL:https://seasevents.nmsdev7.com/event/be-doctoral-dissertation-defense-computational-imaging-and-multiomic-biomarkers-for-precision-medicine-characterizing-heterogeneity-in-lung-cancer-apurva-singh/
LOCATION:Class of 62 Auditorium\, John Morgan Building\, 3620 Hamilton Walk\, Philadelphia\, PA\, 19104
CATEGORIES:Doctoral,Graduate,Student,Dissertation or Thesis Defense
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240213T100000
DTEND;TZID=America/New_York:20240213T113000
DTSTAMP:20260403T173600
CREATED:20240207T163149Z
LAST-MODIFIED:20240207T163149Z
UID:10007848-1707818400-1707823800@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "AI for Antibiotic Discovery"
DESCRIPTION:Computers can be programmed for superhuman pattern recognition of images and text; however\, their application in biology and medicine is still in its infancy. In this talk\, I will discuss our advances over the past half-decade\, which are accelerating discoveries in the crucial and underinvested area of antibiotic discovery. We developed the first antibiotic designed by a computer with proven efficacy in preclinical animal models\, demonstrating that machines and artificial intelligence (AI) could be used to design therapeutic molecules. Our algorithms have accelerated antibiotic discovery\, and for the first time\, we successfully mined the human proteome for antibiotics. Recently\, we expanded our proteome-mining efforts to explore the proteomes of extinct species. Using AI\, my lab discovered the first therapeutic molecules in extinct organisms\, including Neanderthals and Denisovans\, launching the field of molecular de-extinction. Collectively\, our efforts have dramatically reduced the time needed to discover preclinical antibiotic candidates from years to hours. I believe we are on the cusp of a new era in science where advances enabled by AI will help control antibiotic resistance\, infectious disease outbreaks\, and pandemics.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-ai-for-antibiotic-discovery/
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:20240213T110000
DTEND;TZID=America/New_York:20240213T120000
DTSTAMP:20260403T173600
CREATED:20240201T134437Z
LAST-MODIFIED:20240201T134437Z
UID:10007837-1707822000-1707825600@seasevents.nmsdev7.com
SUMMARY:ESE & CIS Spring Seminar - "Beyond the black box: characterizing and improving how neural networks learn"
DESCRIPTION:The predominant paradigm in deep learning practice treats neural networks as “black boxes”. This leads to economic and environmental costs as brute-force scaling remains the performance driver\, and to safety issues as robust reasoning and alignment remain challenging. My research opens up the neural network black box with mathematical and statistical analyses of how networks learn\, and yields engineering insights that improve the efficiency and transparency of these models. In this talk I will present characterizations of (1) how large language models can learn to reason with abstract symbols\, and (2) how hierarchical structure in data guides deep learning\, and will conclude with (3) new tools to distill trained neural networks into lightweight and transparent models.
URL:https://seasevents.nmsdev7.com/event/ese-spring-seminar-tbd-2/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Colloquium
ORGANIZER;CN="Electrical and Systems Engineering":MAILTO:eseevents@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240214T120000
DTEND;TZID=America/New_York:20240214T131500
DTSTAMP:20260403T173600
CREATED:20230928T142325Z
LAST-MODIFIED:20230928T142325Z
UID:10007713-1707912000-1707916500@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: "Enforcing Right to Explanation: Algorithmic Challenges and Opportunities" (Himabindu Lakkaraju\, Harvard University)
DESCRIPTION:ABSTRACT: \nAs predictive and generative models are increasingly being deployed in various high-stakes applications in critical domains including healthcare\, law\, policy and finance\, it becomes important to ensure that relevant stakeholders understand the behaviors and outputs of these models so that they can determine if and when to intervene. To this end\, several techniques have been proposed in recent literature to explain these models. In addition\, multiple regulatory frameworks (e.g.\, GDPR\, CCPA) introduced in recent years also emphasized the importance of enforcing the key principle of “Right to Explanation” to ensure that individuals who are adversely impacted by algorithmic outcomes are provided with an actionable explanation. In this talk\, I will discuss the gaps that exist between regulations and state-of-the-art technical solutions when it comes to explainability of predictive and generative models. I will then present some of our latest research that attempts to address some of these gaps. I will conclude the talk by discussing bigger challenges that arise as we think about enforcing right to explanation in the context of large language models and other large generative models. \nZOOM LINK (if unable to attend in-person): https://upenn.zoom.us/j/94929617939
URL:https://seasevents.nmsdev7.com/event/asset-seminar-himabindu-lakkaraju-harvard-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:20240214T150000
DTEND;TZID=America/New_York:20240214T160000
DTSTAMP:20260403T173600
CREATED:20240124T151237Z
LAST-MODIFIED:20240124T151237Z
UID:10007822-1707922800-1707926400@seasevents.nmsdev7.com
SUMMARY:Spring 2024 GRASP SFI: Fei Miao\, University of Connecticut\, “Learning and Control for Safety\, Efficiency\, and Resiliency of Embodied AI”
DESCRIPTION:This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. \nABSTRACT\nWith rapid evolution of sensing\, communication\, and computation\, integrating learning and control presents significant Embodied AI opportunities. However\, current decision-making frameworks lack comprehensive understanding of the tridirectional relationship among communication\, learning and control\, posing challenges for multi-agent systems in complex environments. In the first part of the talk\, we focus on learning and control with communication capabilities. We design an uncertainty quantification method for collaborative perception in connected autonomous vehicles (CAVs). Our findings demonstrate that communication among multiple agents can enhance object detection accuracy and reduce uncertainty. Building upon this\, we develop a safe and scalable deep multi-agent reinforcement learning (MARL) framework that leverages shared information among agents to improve system safety and efficiency. We validate the benefits of communication in MARL\, particularly in the context of CAVs in challenging mixed traffic scenarios. We incentivize agents to communicate and coordinate with a novel reward reallocation scheme based on Shapley value for MARL. Additionally\, we present our theoretical analysis of robust MARL methods under state uncertainties\, such as uncertainty quantification in the perception modules or worst-case adversarial state perturbations. In the second part of the talk\, we briefly outline our research contributions on robust MARL and data-driven robust optimization for sustainable mobility. We also highlight our research results concerning CPS security. Through our findings\, we aim to advance Embodied AI and CPS for safety\, efficiency\, and resiliency in dynamic environments.
URL:https://seasevents.nmsdev7.com/event/spring-2024-grasp-sfi-fei-miao/
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:20240214T153000
DTEND;TZID=America/New_York:20240214T163000
DTSTAMP:20260403T173600
CREATED:20240116T180348Z
LAST-MODIFIED:20240116T180348Z
UID:10007807-1707924600-1707928200@seasevents.nmsdev7.com
SUMMARY:CBE Seminar: "Systems Engineering for Addressing Critical Challenges in Viral Vector Manufacturing" (Francesco Destro\, MIT)
DESCRIPTION:Abstract\nThe demand for viral vectors is poised to soon exceed current production capacities\, driven by the surging number of clinical trials for gene and cell therapies. Unfortunately\, current manufacturing processes for viral vectors have high costs and low titers. This talk will demonstrate how process systems engineering tools can be leveraged for addressing the most critical challenges in the manufacturing process for recombinant adeno-associated virus (rAAV)\, the most widely used viral vector in commercial gene therapies. FDA recently approved the first rAAV-based gene therapies manufactured in the Sf9/baculovirus expression vector system (BEVS). Within the BEVS\, Sf9 cells produce rAAV as a result of infection with recombinant baculoviruses that carry the genetic blueprint for vector production. A mechanistic model is developed to identify the bottlenecks to full capsid formation in the intracellular pathway for rAAV production in Sf9 cells. The model indicates genetic modifications to the baculovirus vectors that can enhance the productivity of the platform. Further\, the optimal process conditions to establish continuous rAAV manufacturing in the BEVS are identified through a novel numerical method for solving systems of partial differential equations. Finally\, a powerful machine learning model is introduced for real-time prediction of rAAV titers based on single-cell biophysical signatures.
URL:https://seasevents.nmsdev7.com/event/cbe-seminar-systems-engineering-for-addressing-critical-challenges-in-viral-vector-manufacturing-francesco-destro-mit/
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:20240215T110000
DTEND;TZID=America/New_York:20240215T120000
DTSTAMP:20260403T173600
CREATED:20240201T134931Z
LAST-MODIFIED:20240201T134931Z
UID:10007838-1707994800-1707998400@seasevents.nmsdev7.com
SUMMARY:ESE Spring Seminar - "White-Box Computational Imaging: Measurements to Images to Insights"
DESCRIPTION:Computation and machine learning hold tremendous potential to improve the quality and capabilities of imaging methods used across science\, medicine\, engineering\, and art. Despite their impressive performance on benchmark datasets\, however\, deep learning methods are known to behave unpredictably on some real-world data\, which limits their trusted adoption in safety-critical domains. Accordingly\, in this talk I will describe white-box\, interpretable methods for photorealistic volumetric reconstruction that match or exceed the performance of black-box neural alternatives. I will also present recent theoretical results that guarantee correct and efficient reconstruction using our white-box approach in nonlinear computed tomography.
URL:https://seasevents.nmsdev7.com/event/ese-spring-seminar-white-box-computational-imaging-measurements-to-images-to-insights/
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:20240215T153000
DTEND;TZID=America/New_York:20240215T163000
DTSTAMP:20260403T173600
CREATED:20240108T171337Z
LAST-MODIFIED:20240108T171337Z
UID:10007793-1708011000-1708014600@seasevents.nmsdev7.com
SUMMARY:BE Seminar: "Where Do Therapeutic Antibodies Go?: A First-In-Human Journey" (Guolan Lu\, Stanford)
DESCRIPTION:Dr. Lu will introduce a fluorescence molecular imaging method to track therapeutic antibody delivery from cancer patients in vivo\, down to single cells\, through first-in-human clinical trials. She will present a new experimental and AI-powered analytical framework that integrates single-cell drug imaging with spatial omics to decipher drug-target-microenvironment in situ. This work establishes a foundational framework for studying drug pharmacology in the context of tissue biology in serious diseases including cancer\, autoimmunity\, and chronic inflammation.
URL:https://seasevents.nmsdev7.com/event/be-seminar-guolan-lu-stanford/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
CATEGORIES:Student
ORGANIZER;CN="Bioengineering":MAILTO:be@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240215T153000
DTEND;TZID=America/New_York:20240215T163000
DTSTAMP:20260403T173600
CREATED:20240129T171413Z
LAST-MODIFIED:20240129T171413Z
UID:10007833-1708011000-1708014600@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Accessible Foundation Models: Systems\, Algorithms\, and Science"
DESCRIPTION:The ever-increasing scale of foundation models\, such as ChatGPT and AlphaFold\, has revolutionized AI and science more generally. However\, increasing scale also steadily raises computational barriers\, blocking almost everyone from studying\, adapting\, or otherwise using these models for anything beyond static API queries. In this talk\, I will present research that significantly lowers these barriers for a wide range of use cases\, including inference algorithms that are used to make predictions after training\, finetuning approaches that adapt a trained model to new data\, and finally\, full training of foundation models from scratch.  For inference\, I will describe our LLM.int8() algorithm\, which showed how to enable high-precision 8-bit matrix multiplication that is both fast and memory efficient. LLM.int8() is based on the discovery and characterization of sparse outlier sub-networks that only emerge at large model scales but are crucial for effective Int8 quantization. For finetuning\, I will introduce the QLoRA algorithm\, which pushes such quantization much further to unlock finetuning of very large models on a single GPU by only updating a small set of the parameters while keeping most of the network in a new information-theoretically optimal 4-bit representation. For full training\, I will present SWARM parallelism\, which allows collaborative training of foundation models across continents on standard internet infrastructure while still being 80% as effective as the prohibitively expensive supercomputers that are currently used. Finally\, I will close by outlining my plans to make foundation models 100x more accessible\, which will be needed to maintain truly open AI-based scientific innovation as models continue to scale.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-accessible-foundation-models-systems-algorithms-and-science/
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:20240216T103000
DTEND;TZID=America/New_York:20240216T114500
DTSTAMP:20260403T173600
CREATED:20240208T161800Z
LAST-MODIFIED:20240208T161800Z
UID:10007851-1708079400-1708083900@seasevents.nmsdev7.com
SUMMARY:Spring 2024 GRASP on Robotics: Guillaume Sartoretti\, National University of Singapore\, “Towards Learned Cooperation at Scale in Robotic Multi-Agent Systems”
DESCRIPTION:This is a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. \nABSTRACT\nWith the recent advances in sensing\, actuation\, computation\, and communication\, the deployment of large numbers of robots is becoming a promising avenue to enable or speed up complex tasks in areas such as manufacturing\, last-mile delivery\, search-and-rescue\, or autonomous inspection. My group strives to push the boundaries of multi-agent scalability by understanding and eliciting emergent coordination/cooperation in multi-robot systems as well as in articulated robots (where agents are individual joints). Our work mainly relies on distributed (multi-agent) reinforcement learning\, where we focus on endowing agents with novel information and mechanisms that can help them align their decentralized policies towards team-level cooperation. In this talk\, I will first summarize my early work in independent learning\, before discussing my group’s recent advances in convention\, communication\, and context-based learning. I will discuss these techniques within a wide variety of robotic applications\, such as multi-agent path finding\, autonomous exploration/search\, task allocation\, and legged locomotion. Finally\, I will also touch on our recent incursion into the next frontier for multi-robot systems: cooperation learning for heterogeneous multi-robot teams. Throughout this journey\, I will highlight the key challenges surrounding learning representations\, policy space exploration\, and scalability of the learned policies\, and outline some of the open avenues for research in this exciting area of robotics.
URL:https://seasevents.nmsdev7.com/event/spring-2024-grasp-on-robotics-guillaume-sartoretti-national-university-of-singapore-towards-learned-cooperation-at-scale-in-robotic-multi-agent-systems/
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:20240219T140000
DTEND;TZID=America/New_York:20240219T160000
DTSTAMP:20260403T173600
CREATED:20240206T155231Z
LAST-MODIFIED:20240206T155231Z
UID:10007847-1708351200-1708358400@seasevents.nmsdev7.com
SUMMARY:CBE Doctoral Dissertation Defense: "Controlled Deposition of Cargo-Carrying Colloids from Dispersed Liquid to Solid Surfaces" (Paradorn Rummaneethorn)
DESCRIPTION:Abstract: \nGreat advances have been made in encapsulation of (biological) analytes at extremely high throughput via techniques such as microfluidics and/or conjugation. In parallel\, analytical techniques such as mass spectrometry have advanced to analyze biochemical components of complex mixtures with high resolutions. Both capabilities are essential for enabling biology at subcellular scales; rather than engineering a new integrated system possessing both capabilities\, the path of lower resistance may be to bridge such high-throughput encapsulation to high-resolution analytical platforms. However\, existing encapsulation techniques yield colloids in dispersions\, whereas analytical techniques require sample preparation on surfaces. \nIn this work\, we addressed two types of colloids dispersed in liquid medium – liquid droplets and solid particles. For droplets\, we employed a charge injection technique to study the reversible wetting state modulation of water droplets on hydrophobic polydimethylsiloxane (PDMS) surfaces. The system exhibits a high range of wetting modulation (from nonwetting to 20°)\, and we were able to demonstrate two-way cargo transfer between droplet and surface. For dispersed particles\, we employed two techniques to array particles in a patterned microwell array: capillary assembly and dielectrophoretic assembly. For capillary assembly\, we studied the effects of coating speed\, coating passes\, particle concentration\, surface temperature\, and presence of surfactants to optimize yield (% of occupied wells) and selectivity (% of particles inside microwells) of particle arraying. As for dielectrophoretic (DEP) assembly\, we studied the number of particles deposited as a function of peak-to-peak voltage (DEP force) and alternating current frequency (DEP polarity) to the arraying of carboxylate-conjugated polystyrene particles. \nThe physical nature of these technologies enables robustness against combinations of colloid-surface chemical characteristics\, with a tunable parameter space that empowers broad use cases involving different colloid-surface combinations. Beyond the colloid deposition use case described here\, the technologies studied here can also be applied to separations\, heterogeneous reaction engineering\, and fundamental colloid-surface studies. When colloids and surfaces come together\, possibilities are imagination-limited.
URL:https://seasevents.nmsdev7.com/event/cbe-doctoral-dissertation-defense-controlled-deposition-of-cargo-carrying-colloids-from-dispersed-liquid-to-solid-surfaces-paradorn-rummaneethorn/
LOCATION:Towne 337
CATEGORIES:Doctoral,Graduate,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:20240220T100000
DTEND;TZID=America/New_York:20240220T113000
DTSTAMP:20260403T173600
CREATED:20240208T165355Z
LAST-MODIFIED:20240208T165355Z
UID:10007852-1708423200-1708428600@seasevents.nmsdev7.com
SUMMARY:MEAM Seminar: "Precision Engineering in Health and Medicine via Acoustics"
DESCRIPTION:Precision engineering stands at the forefront of catalyzing transformative advancements in health and medicine. Among various precise techniques utilizing phototactic\, electric\, and magnetic mechanisms\, acoustic devices have captured heightened attention for their capability to facilitate contactless\, label-free\, and biocompatible manipulation of cells\, extracellular vesicles\, and organisms. Demonstrating proficiency in manipulating micro/nano-objects across a diverse spectrum\, acoustic platforms have evolved to facilitate cell patterning\, separation\, and sorting for single-cell analysis\, isolate extracellular vesicles for diagnosing cancer and neurodegenerative diseases\, and assemble single cells for tissue engineering. As acoustic research and technologies continue to advance\, acoustic devices emerge as a linchpin\, seamlessly bridging the realms of engineering and medicine. This integration propels the frontier of personalized medicine and advanced manufacturing\, showcasing the transformative potential of precision engineering in shaping the future of healthcare. \nIn this presentation\, I will explore the dynamic application of acoustics to advance precision engineering in health and medicine\, spanning a range from nanometer to millimeter scales. I will showcase a series of noteworthy examples\, including (1) the assembly and dynamic control of colloids\, droplets\, and living cells; (2) the precise separation of extracellular vesicles for disease diagnostics; and (3) the engineering of 3D tissues for therapeutic purposes. The distinctive attributes of acoustic platforms\, such as precision\, biocompatibility\, and versatility\, endow them with immense potential to serve as pioneering technologies\, translating innovations in mechanical engineering into advancements in materials\, biology\, and medicine. Additionally\, I will touch upon my past and ongoing endeavors\, covering topics such as sensors and actuators\, nanofabrication\, and advanced packaging\, showcasing broad applications in the fields of semiconductors\, micro/nanorobotics\, and biodevices.
URL:https://seasevents.nmsdev7.com/event/meam-seminar-precision-engineering-in-health-and-medicine-via-acoustics/
LOCATION:Wu and Chen Auditorium (Room 101)\, Levine Hall\, 3330 Walnut Street\, Philadelphia\, PA\, 19104\, United States
ORGANIZER;CN="Mechanical Engineering and Applied Mechanics":MAILTO:meam@seas.upenn.edu
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240220T153000
DTEND;TZID=America/New_York:20240220T163000
DTSTAMP:20260403T173600
CREATED:20240209T133752Z
LAST-MODIFIED:20240209T133752Z
UID:10007853-1708443000-1708446600@seasevents.nmsdev7.com
SUMMARY:CIS Seminar: "Towards Flexible\, Scalable\, and Knowledgeable Generative Intelligence"
DESCRIPTION:From language modeling to 3D vision\, generative AI has revolutionized nearly every aspect of machine learning. In this talk\, I will examine the limitations of the foundation behind many generative AI techniques–autoregressive models. Despite their impressive successes\, these token-by-token models face various challenges\, including 1). non-flexible computation during generation\, 2). lack of rich inner structures for scalable modeling\, and 3). limited understanding of the real world. \nTo address these three issues\, I propose to strategically predict “latents” for the design of new generative models\, where latents refer to the model’s intermediate representations during the generation process. First\, I will demonstrate how integrating latents allows flexible architecture designs to enhance both efficiency and adaptability \,such as in the first non-autoregressive model for sequence generation. Next\, I will show how to use latents to incorporate useful data structures for improved model scalability\, especially in high-resolution images and videos. Moreover\, I will demonstrate how to use latents to infuse world knowledge such as 3D for tasks like consistent view synthesis. Throughout the talk\, I will cover various modalities\, including text\, images\, and 3D. Finally\, I will conclude with a discussion about the prevailing challenges and envision future paths that could lead to more flexible\, scalable \,a nd knowledgeable next-generation generative models.
URL:https://seasevents.nmsdev7.com/event/cis-seminar-towards-flexible-scalable-and-knowledgeable-generative-intelligence/
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:20240221T120000
DTEND;TZID=America/New_York:20240221T133000
DTSTAMP:20260403T173600
CREATED:20240125T161410Z
LAST-MODIFIED:20240125T161410Z
UID:10007827-1708516800-1708522200@seasevents.nmsdev7.com
SUMMARY:ASSET Seminar: “Mathematical Foundations for Physical Agents” (Max Simchowitz\, Massachusetts Institute of Technology\, CSAIL)
DESCRIPTION:ABSTRACT: \nFrom robotics to autonomous vehicles\, machine learning agents deployed in the physical world (“physical agents”) promise to revolutionize endeavors ranging from manufacturing to agriculture to domestic labor. In this talk\, we will develop mathematical foundations\, from the ground up\, for how to carry out this vision. We will begin our investigation by examining linear dynamical systems\, a simple and fundamental model of the interaction between a physical agent and its environment. We prove mathematically that simple exploration attains optimal performance for some of both the simplest and the most complex learning problems in this class. The above finding\, while powerful\, strongly motivates moving past linear dynamics as a mathematical testbed for understanding learning with physical agents. \nHence\, we turn to providing mathematical guarantees for a setting of real-world importance that does not fit the linear mold: behavior cloning. Behavior cloning — teaching a robot to imitate from example demonstrations — lies at the heart of many of today’s most promising robot learning endeavors due to its intuitive data collection and simplicity. Though it can work incredibly well\, we still do not have a clear understanding of what circumstances ensure its success. Bringing together the flexibility of generative models with key intuitions arising from the study of linear control\,  we introduce a framework for behavior cloning that enables an agent to imitate nearly arbitrary behavior with provable guarantees\, even when the dynamics governing the agent and environments interaction are nonlinear. We conclude by outlining ongoing work and future steps towards building out the mathematical and conceptual tooling for understanding the next steps towards general\, capable and flexible physical agents. \n  \nZOOM LINK (if unable to attend in-person): https://upenn.zoom.us/j/99732583896 \n 
URL:https://seasevents.nmsdev7.com/event/asset-seminar-mathematical-foundations-for-physical-agents-max-simchowitz-massachusetts-institute-of-technology-csail/
LOCATION:Raisler Lounge (Room 225)\, Towne Building\, 220 South 33rd Street\, Philadelphia\, PA\, 19104\, United States
END:VEVENT
END:VCALENDAR