In this talk, I will share some insights and observations from our work on Atlas, the world's most dynamic humanoid robot. I'll talk about some of the core ideas—like model-predictive control (MPC)—that have made an impact for us, how learning is playing a larger role in how we design control systems at Boston Dynamics, and […]
Protecting secrets within computer systems is a central mission of cryptography. Program obfuscation, which scrambles computer code without harming its functionality, is an immensely powerful and versatile tool for accomplishing this task that has been the subject of intense study in classical cryptography. Yet, the ability to obfuscate quantum computation had previously remained elusive to […]
ABSTRACT: Machine learning has been increasingly deployed for high-stakes decisions that deeply impact people’s lives. My research focuses on developing interpretable algorithms and pipelines to ensure the safe and efficient utilization of machine learning models in the decision-making process. In this talk, I will introduce a new paradigm, called learning the Rashomon set, which finds and stores […]
This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. This week's speaker will be virtual. ABSTRACT Autonomous flying robots have become widespread in recent years, yet their capability to interact with the environment. remains limited. Moving in multiple fluids is one of the great challenges of mobile […]
Abstract: In modern engineering, sustainability is a crucial aspect, especially in chemical and biological process systems. This concept has evolved to encompass a holistic approach, emphasizing not only the use of sustainable resources but also the development of systems that are environmentally friendly, smart, and efficient. Such systems aim to minimize waste and optimize resource […]
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 […]
This is a hybrid event with a VIRTUAL SPEAKER. There will be in-person attendance in Wu and Chen and virtual attendance on Zoom. ABSTRACT This 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, […]
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 […]
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. […]
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 […]
This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. ABSTRACT Today'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 […]
ABSTRACT: In 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 […]
Abstract Invasive 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 […]
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.
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, […]
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 […]
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 […]
This is a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. ABSTRACT Generative 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 […]
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 […]
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 […]
The Department of Bioengineering at the University of Pennsylvania and Dr. Despina Kontos are pleased to announce the Doctoral Dissertation Defense of Apurva Singh. Title: "Computational imaging and multiomic biomarkers for precision medicine: characterizing heterogeneity in lung cancer" Date: February 12, 2024 Time: 1:00 PM Location: John Morgan Building, Class of 62 auditorium Zoom link The public is […]
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 […]
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 […]
ABSTRACT: As 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 […]
This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. ABSTRACT With 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 […]
Abstract The 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 […]
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 […]
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 […]
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 […]
This is a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. ABSTRACT With 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, […]
Abstract: Great 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 […]
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, […]
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 […]
ABSTRACT: From 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 […]
This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. This week's speaker will be virtual. There will be an extended Q&A/discussion after the seminar from 4:00 PM to 4:30 PM. ABSTRACT Based on the theory of homogeneous spaces we derive \textit{geometrically optimal edge attributes} to be used […]
Abstract Neuroelectronic interfaces have enabled significant advances in both fundamental neuroscience research and the treatment of neurological disorders. However, current neuroelectronic devices have a clear trade-off between invasiveness and spatial resolution, and are unable to achieve seamless integration into the nervous system with cell-type specificity. In this talk, I will first introduce an ultra-small and […]
Traditional biomedical imaging techniques face throughput bottlenecks that limit our ability to study complex dynamic samples like cells, organoids, tissues, and organisms. In particular, hardware-only systems have inherent physical limitations preventing the simultaneous improvement of resolution, field of view, and frame rate. In this seminar, I propose that large-scale, machine learning-accelerated computational imaging will be […]
In this seminar, I explore the pivotal role of tissue stem cells and their microenvironment, known as the niche, in regeneration. With a focus on the endothelial cell niche, my work introduces an innovative in vitro vascularized perfusable model tailored for human tissue explants and organoids. My findings also highlight the crucial contribution of paracrine […]
This is a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. ABSTRACT Magnetic fields offer the possibility of manipulating objects from a distance and are ideal for medical applications, as they penetrate human tissue without inflicting any harm on the patient. Magnetic fields can be harnessed to actuate surgical robots, […]
Technology is dramatically driving down the cost of sequencing genetic data and increasing the quality of predictions made with this data. A standard concern is that these predictions could impair the functioning of insurance markets, either because insurers would abuse genetic information or because of adverse selection. We make three contributions. First, we develop a […]
Abstract: This thesis addresses the fundamental questions surrounding the design and functional capabilities of enzyme-powered micromotors synthesized using microfluidic techniques. The research focuses on the development of these motors, made from artificial cell (protocell) scaffolds, and which seek to replicate the motion behavior of biological cells, investigating their propulsion mechanisms, motion directionality, and collective behavior. […]
This presentation will explore the critical process of curating medical imaging data for AI algorithm development in ophthalmology, highlighting the challenges and current limitations in data curation. It will discuss benchmark datasets, reference standards for FDA validation, and innovative strategies to enhance data availability. Attendees will gain insights into best practices and future directions in […]
Engineered, personalized tissues are transforming healthcare by accelerating clinical testing and serving as cell-based therapies. The emergence of induced pluripotent stem cells coupled with innovations in biomanufacturing and cellular engineering have rapidly advanced engineered tissue cellular and structural complexity. However, generating functional, mature tissues remains an enormous challenge. Control over biophysical forces governing development and […]
Machine learning models trained on vast amounts of data have achieved remarkable success across various applications. However, they also pose new challenges and risks for deployment in real-world high-stakes domains. Decisions made by deep learning models are often difficult to interpret, and the underlying mechanisms remain poorly understood. Given that deep learning models operate as […]
Machine learning models pretrained on internet data have acquired broad knowledge about the world but struggle to solve complex tasks that require extended reasoning and planning. Sequential decision making, on the other hand, has empowered AlphaGo’s superhuman performance, but lacks visual, language, and physical knowledge about the world. In this talk, I will present my […]
ABSTRACT: The rapid advance of AI driven by Large Language Models (LLMs), like ChatGPT, has led to impressive results across a range of different use cases. This has included several models developed for the medical domain which have exhibited surprising behaviors, such as answering medical questions and performing well on medical licensing exams. These results have demonstrated the coming transformation of medicine by […]
This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. This week's speaker will be virtual. ABSTRACT Various applications for autonomous multi-robot systems, such as disaster response, infrastructure repair, agricultural operations, contaminant cleanup, and wildfire suppression, will require these systems to act as distributed sensors, sources, and manipulators […]
Abstract DNA-binding proteins (DBPs) play critical roles in biology and biotechnology, and there has been considerable interest in the engineering of DBPs with new functions or altered specificities. While there has been success in reprogramming and the specificity of naturally occurring DBPs using selection methods, the computational design of new DBPs that engage with DNA […]
Classical and quantum light sources play a fundamental role in science and technology from quantum computing, to communications, manufacturing, defense, sensing, medicine, or imaging. However, scaling the power of lasers has always come at the cost of single mode operation, a scaling question that has been investigated, without success, since the invention of lasers in […]
From quantum physics to cosmology, researchers aim to see things which are typically invisible – be it the entanglement of two particles or infrared signatures from space. In these and various other fields, we are confronted by a common challenge: What we can see with our own eyes or observe using standard optical imaging systems […]
In many computer systems today, an attacker that breaks one system component can steal data from millions of users. In this talk, I will present three systems that can withstand component compromise. I will describe (1) a single sign-on system that protects user security and privacy from a compromised single sign-on server, (2) a secure-hardware-based […]
This is a hybrid event with in-person attendance in Wu and Chen and virtual attendance on Zoom. This seminar will NOT be recorded. ABSTRACT Modeling 3D objects and scenes creates many challenges, both on the analysis and the synthesis sides. The focus of this talk is the compositional structure of objects (into parts) and of […]
Learn how to commercialize your research and navigate Penn's innovation ecosystem with Penn Engineering: Entrepreneurship. This is the first part of a series from Penn Engineering’s Entrepreneurship group. Featuring: Andrew Tsourkas, Professor of Bioengineering, Co-Founder, AlphaThera, Inc, Co-Founder, StreamLaunch, Llc and Jeffrey Babin, Professor of Practice and Associate Director of Entrepreneurship at Penn Engineering & […]
