Universal approximation of input-output maps by temporal convolutional nets

November 22, 2019 - 11:15 AM TSRB Auditorium Max Raginsky University of Illinois at Urbana Champaign Abstract A number of problems in machine learning involve sequence-to-sequence transformations, i.e., nonlinear operators that map an input sequence to an output sequence. Traditionally, such input-output maps have been modeled using discrete-time recurrent neural nets. However, there has been … Continue reading Universal approximation of input-output maps by temporal convolutional nets

Formal Verification of End-to-End Deep Reinforcement Learning

November 08, 2019 - 11:15 AM TSRB Auditorium Yasser Shoukry University of California, Irvine Abstract From simple logical constructs to complex deep neural network models, Artificial Intelligence (AI)-agents are increasingly controlling physical/mechanical systems. Self-driving cars, drones, and smart cities are just examples of such systems to name a few. However, regardless of the explosion in … Continue reading Formal Verification of End-to-End Deep Reinforcement Learning

How I Learned to Stop Worry and Start Loving Lifting to Infinite Dimensions

November 01, 2019 - 11:15 AM TSRB Auditorium Ram Vasudevan University of Michigan Abstract Autonomous systems offer the promise of providing greater safety and access. However, this positive impact will only be achieved if the underlying algorithms that control such systems can be certified to behave robustly. This talk will describe a pair of techniques … Continue reading How I Learned to Stop Worry and Start Loving Lifting to Infinite Dimensions

Monitoring Over the Long Term and Rethinking Interaction in Multi-Robot Teams

October 25, 2019 - 11:15 AM TSRB Auditorium Ryan Williams Virginia Tech Abstract In the area of multi-robot systems, an understanding of collaboration has been developed and is showing great promise in real-world applications including self-driving cars, industrial automation, precision agriculture, and disaster response. However, existing multi-robot methods still leave much to be desired. In … Continue reading Monitoring Over the Long Term and Rethinking Interaction in Multi-Robot Teams

Low Gain Feedback: For Constrained Control, Nonlinear Stabilization and Control of Time Delay Systems

October 11, 2019 - 11:15 AM TSRB 523A Zongli Lin University of Virginia Abstract Low gain feedback refers to a family of stabilizing state feedback gains that are parameterized in a scalar, referred to low gain parameter, and go to zero as the low gain parameter decreases to zero. Low gain feedback was initially proposed … Continue reading Low Gain Feedback: For Constrained Control, Nonlinear Stabilization and Control of Time Delay Systems

Distributed Energy Resources: PDEs and Hopfield Methods

October 04, 2019 - 11:15 AM TSRB Auditorium Scott Moura UC Berkeley Abstract Variable renewable energy integration and resilience to extreme events motivate the need for flexible resources in electric power systems. Distributed energy resources (DERs), such as electric vehicles and thermostatically controlled loads, provide an intriguing set of distributed assets to provide flexible services … Continue reading Distributed Energy Resources: PDEs and Hopfield Methods

A Geometric Method of Hoverability Analysis for Multirotor UAVs

September 27, 2019 - 11:15 AM TSRB Auditorium Tatsuya Ibuki Tokyo Institute of Technology Abstract This talk presents a novel geometric method to investigate whether a multirotor unmanned aerial vehicle (UAV) can achieve stable hovering, i.e., hoverability. The hoverability is indispensable for a multirotor UAV to conduct its task safely, and should be satisfied even … Continue reading A Geometric Method of Hoverability Analysis for Multirotor UAVs