CMChristos N. Mavridis

Christos N. Mavridis

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Christos N. Mavridis has a total of 11 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Learning a network digital twin as a hybrid system

    by Christos Mavridis, Fernando S. Barbosa, Hamed Farhadi, Karl H. Johansson


  • Open Access

    Real-Time Switched System Identification With Online Deterministic Annealing

    by Christos N. Mavridis, Karl Henrik Johansson

    We introduce a real-time identification method for discrete-time state-dependent switching systems in both the input–output and state-space domains. In particular, we design a system of adaptive algorithms running in two timescales; a stochastic approximation algorithm implements an online deterministic annealing scheme at a slow timescale and estimates the mode-switching signal, and a recursive identification algorithm runs at a faster timescale and updates the parameters of the local models based on the estimate of the switching signal. We first focus on piece-wise affine systems and discuss identifiability conditions and convergence properties based on the theory of two-timescale stochastic approximation. In contrast to standard identification algorithms for switched systems, the proposed approach gradually estimates the number of modes and is appropriate for real-time system identification using sequential data acquisition. The progressive nature of the algorithm improves computational efficiency and provides real-time control over the performance-complexity trade-off. Finally, we address specific challenges that arise in the application of the proposed methodology in identification of more general switching systems. Simulation results validate the efficacy of the proposed methodology.


  • Risk-Sensitive Reinforcement Learning With Exponential Criteria

    by Erfaun Noorani, Christos N. Mavridis, John S. Baras

    While reinforcement learning (RL) has shown experimental success in a number of applications, it is known to be sensitive to noise and perturbations in the parameters of the system, leading to high variability in the total reward amongst different episodes on slightly different environments. To introduce robustness, as well as sample efficiency, risk-sensitive RL methods are being thoroughly studied. In this work, we provide a definition of robust RL policies and formulate a risk-sensitive RL problem to approximate them, by solving an optimization problem with respect to a modified objective based on exponential criteria. In particular, we study a model-free risk-sensitive variation of the widely used Monte Carlo policy gradient algorithm, and introduce a novel risk-sensitive online Actor-Critic algorithm based on solving a multiplicative Bellman equation using stochastic approximation updates. Analytical results suggest that the use of exponential criteria generalizes commonly used ad-hoc regularization approaches, improves sample efficiency, and introduces robustness with respect to perturbations in the model parameters and the environment. The implementation, performance, and robustness properties of the proposed methods are evaluated in simulated experiments.


  • Open Access

    Learning Agent Interactions from Density Evolution in 3D Regions With Obstacles

    by Amoolya Tirumalai, Christos N. Mavridis, John S. Baras

    In this work, we study the inverse problem of identifying complex flocking dynamics in a domain cluttered with obstacles. We get inspiration from animal flocks moving in complex ways with capabilities far beyond what current robots can do. Owing to the difficulty of observing and recovering the trajectories of the agents, we focus on the dynamics of their probability densities, which are governed by partial differential equations (PDEs), namely compressible Euler equations subject to non-local forces. We formulate the inverse problem of learning interactions as a PDE-constrained optimization problem of minimizing the squared Hellinger distance between the histogram of the flock and the distribution associated to our PDEs. The numerical methods used to efficiently solve the PDE-constrained optimization problem are described. Realistic flocking data are simulated using the Boids model of flocking agents, which differs in nature from the reconstruction models used in our PDEs. Our analysis and simulated experiments show that the behavior of cohesive flocks can be recovered accurately with approximate PDE solutions.


  • Open Access

    Weak Solutions to an Euler Alignment System with Singular Interactions in a Bounded Domain

    by Amoolya Tirumalai, Christos Mavridis, John S. Baras

    Euler alignment systems appear as hydrodynamic limits of interacting self-propelled particle systems such as the (generalized) Cucker-Smale model. In this work, we study weak solutions to an Euler alignment system on smooth, bounded, connected domains. This particular Euler alignment system includes singular alignment, attraction, and repulsion interaction kernels which correspond to a Yukawa potential. We also include a confinement potential and self-propulsion. We embed the problem into an abstract Euler system to conclude that infinitely many weak solutions exist. We further show that we can construct solutions satisfying bounds on an energy quantity, and that the solutions satisfy a weak-strong uniqueness principle. Finally, we present an addition of leader-agents governed by controlled ODEs, and modification of the interactions to be Bessel potentials of fractional order $s > 2$.

  • Annealing Optimization for Progressive Learning With Stochastic Approximation

    by Christos N. Mavridis, John S. Baras

    In this work, we introduce a learning model designed to meet the needs of applications in which computational resources are limited, and robustness and interpretability are prioritized. Learning problems can be formulated as constrained stochastic optimization problems, with the constraints originating mainly from model assumptions that define a trade-off between complexity and performance. This trade-off is closely related to over-fitting, generalization capacity, and robustness to noise and adversarial attacks, and depends on both the structure and complexity of the model, as well as the properties of the optimization methods used. We develop an online prototype-based learning algorithm based on annealing optimization that is formulated as an online gradient-free stochastic approximation algorithm. The learning model can be viewed as an interpretable and progressively growing competitive-learning neural network model to be used for supervised, unsupervised, and reinforcement learning. The annealing nature of the algorithm contributes to minimal hyper-parameter tuning requirements, poor local minima prevention, and robustness with respect to the initial conditions. At the same time, it provides online control over the performance-complexity trade-off by progressively increasing the complexity of the learning model as needed, through an intuitive bifurcation phenomenon. Finally, the use of stochastic approximation enables the study of the convergence of the learning algorithm through mathematical tools from dynamical systems and control, and allows for its integration with reinforcement learning algorithms, constructing an adaptive state-action aggregation scheme.


  • Learning Swarm Interaction Dynamics From Density Evolution

    by Christos N. Mavridis, Amoolya Tirumalai, John S. Baras

    We focus on understanding the coordinated movements of animal or artificial swarms, and propose a learning scheme to capture the coordination laws of the interacting agents, from observations of the swarm's density over time. We describe the dynamics of the interacting agents with a general Cucker-Smale flocking model, and propose a family of parametric interaction functions which allows the mean-field macroscopic system of partial integro-differential equations, describing the swarm's density, to be efficiently solved as an augmented system of PDEs. We solve the augmented system in an iterative optimization scheme to learn the dynamics of the interacting agents from observations of the swarm's density evolution. The results of this work can give insights on how animal flocks coordinate, give rise to new control schemes for large distributed systems, and serve as a central part of defence mechanisms against adversarial UAV swarms. We illustrate our methodology using simulation data in two dimensions.


  • Online Deterministic Annealing for Classification and Clustering

    by Christos N. Mavridis, John S. Baras

    Inherent in virtually every iterative machine learning algorithm is the problem of hyperparameter tuning, which includes three major design parameters: 1) the complexity of the model, e.g., the number of neurons in a neural network; 2) the initial conditions, which heavily affect the behavior of the algorithm; and 3) the dissimilarity measure used to quantify its performance. We introduce an online prototype-based learning algorithm that can be viewed as a progressively growing competitive-learning neural network architecture for classification and clustering. The learning rule of the proposed approach is formulated as an online gradient-free stochastic approximation algorithm that solves a sequence of appropriately defined optimization problems, simulating an annealing process. The annealing nature of the algorithm contributes to avoiding poor local minima, offers robustness with respect to the initial conditions, and provides a means to progressively increase the complexity of the learning model, through an intuitive bifurcation phenomenon. The proposed approach is interpretable, requires minimal hyperparameter tuning, and allows online control over the performance-complexity tradeoff. Finally, we show that Bregman divergences appear naturally as a family of dissimilarity measures that play a central role in both the performance and the computational complexity of the learning algorithm.


  • Vector Quantization for Adaptive State Aggregation in Reinforcement Learning

    by Christos N. Mavridis, John S. Baras

    We propose an adaptive state aggregation scheme to be used along with temporal-difference reinforcement learning and value function approximation algorithms. The resulting algorithm constitutes a two-timescale stochastic approximation algorithm with: (a) a fast component that executes a temporal-difference reinforcement learning algorithm, and (b) a slow component, based on online vector quantization, that adaptively partitions the state space of a Markov Decision Process according to an appropriately defined dissimilarity measure. We study the convergence of the proposed methodology using Bregman Divergences as dissimilarity measures that can increase the efficiency and reduce the computational complexity of vector quantization algorithms. Finally, we quantify its performance on the Cart-pole (inverted pendulum) optimal control problem using Q-learning with adaptive state aggregation based on the Self-Organizing Map (SOM) algorithm.