5 papers
Walking the Values in Bayesian Inverse Reinforcement Learning
Ondrej Bajgar, Alessandro Abate, Konstantinos Gatsis +1
The goal of Bayesian inverse reinforcement learning (IRL) is recovering a posterior distribution over reward functions using a set of demonstrations from an expert optimizing for a…
Scalable Forward Reachability Analysis of Multi-Agent Systems with Neural Network Controllers
Oliver Gates, Matthew Newton, Konstantinos Gatsis
Neural networks (NNs) have been shown to learn complex control laws successfully, often with performance advantages or decreased computational cost compared to alternative methods.…
Homomorphically encrypted gradient descent algorithms for quadratic programming
André Bertolace, Konstantinos Gatsis, Kostas Margellos
In this paper, we evaluate the different fully homomorphic encryption schemes, propose an implementation, and numerically analyze the applicability of gradient descent algorithms t…
Large-Scale Graph Reinforcement Learning in Wireless Control Systems
Vinicius Lima, Mark Eisen, Konstantinos Gatsis +1
Modern control systems routinely employ wireless networks to exchange information between spatially distributed plants, actuators and sensors. With wireless networks defined by ran…
Federated Reinforcement Learning at the Edge
Konstantinos Gatsis
Modern cyber-physical architectures use data collected from systems at different physical locations to learn appropriate behaviors and adapt to uncertain environments. However, an…