Loading view.
Graduate
ESE Seminar: “The Role of Explicit Regularization in Overparameterized Neural Networks”
Zoom - Email ESE for Link jbatter@seas.upenn.eduOverparameterized neural networks have proved to be remarkably successful in many complex tasks such as image classification and deep reinforcement learning. In this talk, we will consider the role of explicit regularization in training overparameterized neural networks. Specifically, we consider ReLU networks and show that the landscape of commonly used regularized loss functions have the […]
ESE Seminar: “Learning is Pruning”
Zoom - Email ESE for Link jbatter@seas.upenn.eduThe strong lottery ticket hypothesis (LTH) postulates that any neural network can be approximated by simply pruning a sufficiently larger network of random weights. Recent work establishes that the strong LTH is true if the random network to be pruned is a large poly-factor wider than the target one. This polynomial over-parameterization is at odds with […]
CBE PhD Dissertation Defense | Investigation of High-Surface-Area Titanate (ATiO3) Thin Films Prepared by Atomic Layer Deposition
Zoom - Email CBE for linkAbstract: "Heterogeneous catalysis is critically important in the chemical-processing and energy-related industries. Ideally, reaction sites on heterogeneous catalysts can be regenerated, and the catalysts should be active throughout the catalytic process. However, harsh reaction conditions and parallel side reactions can be harmful to the metal catalysts, making deactivation a common phenomenon. Finding a catalyst that […]
CBE PhD Dissertation Defense | Study of Metal Catalysts Supported on Thin Films of Perovskites prepared by Atomic Layer Deposition (ALD)
Zoom - Email CBE for linkAbstract: Sintering is a severe problem with supported-metal catalysts in high-temperature applications, such as in automotive-emissions control, because it leads to a loss of catalytically active surface area. To stabilize the metal particles, in 2001, the Daihatsu group proposed using perovskite-supported metals, which they referred to as “intelligent” catalysts. The original goal was to regenerate […]
CBE PhD Dissertation Defense | Exploring the Molecular Origins of Icephilicity Using Specialized Molecular Simulations
Zoom - Email CBE for linkAbstract: Exercising control over the formation of ice and similar crystalline structures is important in a variety of contexts, from preserving organs for transplant to preventing clathrate hydrate plugs in natural gas pipelines. To achieve this control, it is crucial to understand nucleation phenomena at the molecular level. Studies have shown that heterogeneous nucleation proceeds […]
CBE PhD Dissertation Defense | Establishing the Transient Mass Balance of Thrombosis Under Venous Flow: From A Microfluidic Approach To A Reduced Model
Zoom - Email CBE for linkAbstract: Coagulation kinetics are well established in well plates assays in which human plasma clots isotropically. However, less is known about thrombin kinetics and transport within clots formed under hemodynamic flow. Using microfluidic perfusion of Factor XIIa-inhibited human whole blood over a 250-micron long patch of collagen/tissue factor and immunoassays of the effluent for fragment […]
CBE PhD Dissertation Defense | A Computational Approach to the Study of Trauma
Zoom - Email CBE for linkAbstract: Trauma 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 […]
CBE PhD Dissertation Defense | A Scalable, Point-of-Care Microfluidic Approach for Assessing Thrombosis and Hemostasis
Zoom - Email CBE for linkAbstract: Coagulation testing is an important diagnostic tool for the detection of excessive bleeding risk or obstructive clot formation (thrombosis), using blood samples from patients. Microfluidic flow devices have been well established to provide insights on the impacts of shear rate, drug action, and disease state on coagulation and platelet biology. The bulk of the […]
BE Dissertation: “Large Scale Integration of Microengineered Tissue Models for High-content, High-throughput analysis of Complex Human Physiological Systems”
https://upenn.zoom.us/j/96715197752The Department of Bioengineering at the University of Pennsylvania and Dr. Dan Huh are pleased to announce the Doctoral Dissertation Defense of Andrei Georgescu. Title: Large scale integration of microengineered tissue models for high-content, high-throughput analysis of complex human physiological systems. Date: Tuesday April 20, 2021 Time: 2:00pm You are invited to attend via zoom: Zoom […]