Damian Owerko

I am a PhD student and research fellow at the University of Pennsylvania in the department of Electrical and Systems Engineering. There, I received my Master’s in Robotics, B.A. in Physics, and B.S.E. in Systems Science and Engineering. I study under Professor Alejandro Ribeiro and expect to complete my PhD in 2025.

My research program is focused on scalable learning for (distributed) cyber-physical systems. This is a broad class of systems, including power systems, communication systems, and sensor networks. There are various optimization problems such as optimal power flow, wireless routing, and multi-target tracking. Traditional optimization approaches often do not scale well. In contrast, machine learning approaches can often obtain solutions quickly, but require large amounts of data and time to train. Furthermore, they must be trained from scratch to adapt to new scenarios. My goal is to develop learning algorithms that leverage transfer learning to quickly adapt to new problems.

news

Sep 22, 2023	Multi-Target Tracking with Transferable Convolutional Neural Networks was accepted to CAMSAP 2023.
Jul 14, 2023	Solving Large-scale Spatial Problems with Convolutional Neural Networks was accepted to ACSSC 2023.

selected publications

2023

CAMSAP

Multi-Target Tracking with Transferable Convolutional Neural Networks

Damian Owerko, Charilaos Kanatsoulis, Alejandro Ribeiro, and 2 more authors

In 2023 9th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) Jul 2023

Abs Code

Multi-target tracking (MTT) is a traditional signal processing task, where the goal is to estimate the states of an unknown number of moving targets from noisy sensor measurements. In this paper, we revisit MTT from a deep learning perspective and propose convolutional neural network (CNN) architectures to tackle it. We represent the target states and sensor measurements as images. Thereby we recast the problem as a image-to-image prediction task for which we train a fully convolutional model. This architecture is motivated by a novel theoretical bound on the transferability error of CNN. The proposed CNN architecture outperforms a GM-PHD filter on the MTT task with 10 targets. The CNN performance transfers without re-training to a larger MTT task with 250 targets with only a \13\% increase in average OSPA.
TSP

Transferability of Convolutional Neural Networks in Stationary Learning Tasks

Damian Owerko, Charilaos I. Kanatsoulis, Jennifer Bondarchuk, and 2 more authors

Jul 2023

Abs Code

Recent advances in hardware and big data acquisition have accelerated the development of deep learning techniques. For an extended period of time, increasing the model complexity has led to performance improvements for various tasks. However, this trend is becoming unsustainable and there is a need for alternative, computationally lighter . In this paper, we introduce a novel framework for efficient training of convolutional neural networks (CNNs) for large-scale spatial problems. To accomplish this we investigate the properties of CNNs for tasks that the underlying signals are stationary and exhibit translational symmetries. We show that a CNN trained on small windows of such signals achieves a nearly identical zero-shot performance on much larger windows. This claim is supported by our theoretical analysis, which provides a bound on the performance degradation. Additionally, we conduct thorough experimental analysis on two tasks: multi-target tracking and mobile infrastructure on demand. Our results show that the CNN is able to tackle problems with many hundreds of agents after being trained with fewer than ten. Thus, CNN architectures provide solutions to these problems at previously computationally intractable scales.

2022

ICASSP

Unsupervised Optimal Power Flow Using Graph Neural Networks

Damian Owerko, Fernando Gama, and Alejandro Ribeiro

Oct 2022

Abs Code

Optimal power flow (OPF) is a critical optimization problem that allocates power to the generators in order to satisfy the demand at a minimum cost. Solving this problem exactly is computationally infeasible in the general case. In this work, we propose to leverage graph signal processing and machine learning. More specifically, we use a graph neural network to learn a nonlinear parametrization between the power demanded and the corresponding allocation. We learn the solution in an unsupervised manner, minimizing the cost directly. In order to take into account the electrical constraints of the grid, we propose a novel barrier method that is differentiable and works on initially infeasible points. We show through simulations that the use of GNNs in this unsupervised learning context leads to solutions comparable to standard solvers while being computationally efficient and avoiding constraint violations most of the time.