ESE PhD Thesis Defense: “Algorithms for Adversarially Robust Deep Learning”

Wu and Chen Auditorium (Room 101), Levine Hall 3330 Walnut Street, Philadelphia, PA, United States

Given the widespread use of deep learning models in safety-critical applications, ensuring that the decisions of such models are robust against adversarial exploitation is of fundamental importance.  In this thesis, we discuss recent progress toward designing algorithms that exhibit desirable robustness properties.  First, we discuss the problem of adversarial examples in computer vision, for which […]

GRASP Seminar: Lillian Chin, University of Texas at Austin, “Materials Make the Bot: Directly Embedding Actuation and Perception into Robotic Structures”

Levine 307 3330 Walnut Street, Philadelphia, PA, United States

*This seminar will be held in-person in Levine 307 as well as virtually via Zoom.  ABSTRACT To make a future where robots are helpful and commonplace, robots must physically interact with humans and their surroundings. In this talk, I argue that robots should be designed from a materials-centric approach to better facilitate these interactions. If […]

xLab Seminar: “Learning to Control with Vision–Language Models”

Towne 337

If learning from data is valuable, can learning from big data be very valuable? It has been, so far, in vision and language, for which foundation models can be trained on web-scale data to support a plethora of downstream tasks; not so much in control, for which scalable learning remains elusive. Can information encoded in […]

IDEAS Seminar: “An optimization framework for designing robust state estimators”

Levine 307 3330 Walnut Street, Philadelphia, PA, United States

Cyber-physical systems often include communication networks that ensure data transmission between different components of the system (sensors, actuators, processing units, etc). The presence of such networks renders the whole system vulnerable to malicious attacks consisting, for example, in the injection of arbitrary signals. In this context, the data collected over the communication channel may be […]

BE Doctoral Dissertation Defense: “Goal-Directed Dynamics of Network Topology” (Shubhankar Patankar)

Heilmeier Hall (Room 100), Towne Building 220 South 33rd Street, Philadelphia, PA, United States

The Department of Bioengineering at the University of Pennsylvania and Dr. Dani Smith Bassett are pleased to announce the Doctoral Dissertation Defense of Shubhankar Patankar. Title: Goal-Directed Dynamics of Network Topology Date: June 7, 2024 Time: 11:30 am - 1:30 pm Location: Heilmeier Hall, Room 100 Towne Building Zoom: https://upenn.zoom.us/j/95495692170 The public is welcome to attend.

infoLeNS Seminar: “Emerging computational imaging inverse problems: from theory to algorithms”

Room 452 C, 3401 Walnut 3401 Walnut Street, Philadelphia, PA, United States

In this talk, I will focus on two challenging imaging systems: snapshot compressive imaging and coherent imaging under speckle noise interference. I will begin by reviewing the core mathematical modeling of the inverse problem corresponding to each system. I will develop a maximum likelihood estimator (MLE)-based optimization for each, employing untrained neural networks (NNs) to […]

Community for Rigor Public Talk and Live Podcast Recording: “Rigor in Plain English”

Tedori Family Auditorium, Levin Building 425 S University Ave, Philadelphia, PA, United States

You’ve probably seen reports about reproducibility issues and science gone wrong due to lack of rigor. But what does it mean to do rigorous scientific research? If the word conjures up images of stiffness, strictness, and difficulty, the Community for Rigor is here to help! Join us for Rigor in Plain English, a live podcast […]

GRASP Seminar: Matthias Nießner, Technical University of Munich, “Photo-realistic AI Avatars”

Levine 307 3330 Walnut Street, Philadelphia, PA, United States

*This seminar will be held in-person in Levine 307 as well as virtually via Zoom.  ABSTRACT In this keynote, I will talk about our latest research on creating photo-realistic AI Avatars. Here, our main goal is to create virtual characters that can are visually indistinguishable from photos and videos of real people. Further, we aim […]

GRASP Seminar: Angela Dai, Technical University of Munich, “From Understanding to Interacting with the 3D World”

Levine 307 3330 Walnut Street, Philadelphia, PA, United States

*This seminar will be held in-person in Levine 307 as well as virtually via Zoom.  ABSTRACT Understanding the 3D structure of real-world environments is a fundamental challenge in machine perception, critical for applications spanning robotic navigation, content creation, and mixed reality scenarios. In recent years, machine learning has undergone rapid advancements; however, in the 3D […]

MEAM Ph.D. Thesis Defense: “Bistable Structures Enable Passive Transitions in Mobile Robots”

Towne 319 220 S. 33rd Street, Philadelphia, United States

Making robots more capable, agile, and efficient requires careful design of the robot’s mechanical body to match task requirements. Passive components allow a robot to perform a task without a dedicated actuator, often improving both power consumption and overall performance. In this thesis, we investigate robotic applications of bistable mechanisms, mechanical structures that exhibit two […]