Jorge I. Poveda

We introduce a class of concurrent learning (CL) algorithms designed to solve parameter estimation problems with convergence rates ranging from hyperexponential to prescribed-time while utilizing alternating datasets during the learning process. The proposed algorithm employs a broad class of dynamic gains, from exponentially growing to finite-time blow-up gains, enabling either enhanced convergence rates or user-prescribed convergence time independent of the dataset's richness. The CL algorithm can handle applications involving switching between multiple datasets that may have varying degrees of richness and potential corruption. The main result establishes convergence rates faster than any exponential while guaranteeing uniform global ultimate boundedness in the presence of disturbances, with an ultimate bound that shrinks to zero as the magnitude of measurement disturbances and corrupted data decreases. The stability analysis leverages tools from hybrid dynamical systems theory, along with a dilation/contraction argument on the hybrid time domains of the solutions. The algorithm and main results are illustrated via a numerical example.

We study the problem of online feedback optimization for stochastic linear plants and quadratic performance costs in the presence of an unknown disturbance and persistent white noise. For this problem, we introduce a novel two-timescale control architecture consisting of a static feedback term and a dynamically tuned feedforward term. Through singular perturbation analysis, we show that the steady-state mean of the system converges to the feasible optimizer of the problem, independently of the feedback term, and the steady-state covariance converges to the standard LQR baseline up to a small bias of order <inline-formula> <tex-math notation="LaTeX">$\mathcal {O}(\varepsilon)$ </tex-math></inline-formula>, where <inline-formula> <tex-math notation="LaTeX">$\varepsilon $ </tex-math></inline-formula> is the learning rate of the optimization dynamics. In this sense, our results highlight a novel separation principle that justifies a sequential design approach for stochastic feedback optimization: the feedback term may be designed offline via LQR techniques to minimize the covariance penalty, e.g., using a Riccati equation, while the dynamic feedforward leverages a gradient-based architecture that rejects the unknown disturbance from the steady-state mean, all while incurring only an <inline-formula> <tex-math notation="LaTeX">$\mathcal {O}(\varepsilon)$ </tex-math></inline-formula>- optimality gap. Numerical simulations are provided to validate the theory.

This paper introduces a novel Lyapunov-based small-gain methodology for establishing fixed-time stability (FxTS) guarantees in interconnected dynamical systems. Specifically, we consider interconnections in which each subsystem admits an individual fixed-time input-to-state stability (ISS) Lyapunov function that certifies FxT-ISS. We then show that if a nonlinear small-gain condition is satisfied, then the entire interconnected system is FxTS. Our results are analogous to existing Lyapunov-based small-gain theorems developed for asymptotic and finite-time stability, thereby filling an important gap in the stability analysis of interconnected dynamical systems. The proposed theoretical tools are further illustrated through analytical and numerical examples, including the first result on fixed-time feedback optimization of dynamical systems without time-scale separation between the plant and the controller.

Deception is a key tactic for agents in adversarial environments, used to mislead opponents into adopting unaware strategies. In cyber-physical systems, for instance, deception can conceal attacks against critical infrastructure. This tutorial highlights the usefulness of deception for attacking and protecting systems against adversaries, but also as a tool to increase payoff in general game-theoretic and data-driven settings. It presents several state-of-the-art techniques for control-theoretic deception, including deception in defensive cyber-physical security, game-theoretic reinforcement learning, general multi-agent learning systems, Nash equilibrium seeking, and data-driven control. Although showcased in specific contexts, the underlying concepts and ideas that we study should be generalizable by researchers to settings beyond the scope of this tutorial.

This paper studies a power transmission system with both conventional generators (CGs) and distributed energy assets (DEAs) providing frequency control. We consider an operating condition with demand aggregating two dynamic components: one that switches between different values on a finite set, and one that varies smoothly over time. Such dynamic operating conditions may result from protection scheme activations, external cyber-attacks, or due to the integration of dynamic loads, such as data centers. Mathematically, the dynamics of the resulting system are captured by a system that switches between a finite number of vector fields -- or modes--, with each mode having a distinct equilibrium point induced by the demand aggregation. To analyze the stability properties of the resulting switching system, we leverage tools from hybrid dynamic inclusions and the concept of $Ω$-limit sets from sets. Specifically, we characterize a compact set that is semi-globally practically asymptotically stable under the assumption that the switching frequency and load variation rate are sufficiently slow. For arbitrarily fast variations of the load, we use a level-set argument with multiple Lyapunov functions to establish input-to-state stability of a larger set and with respect to the rate of change of the loads. The theoretical results are illustrated via numerical simulations on the IEEE 39-bus test system.

In the theory of multi-agent systems, deception refers to the strategic manipulation of information to influence the behavior of other agents, ultimately altering the long-term dynamics of the entire system. Recently, this concept has been examined in the context of model-free Nash equilibrium seeking (NES) algorithms for noncooperative games. Specifically, it was demonstrated that players can exploit knowledge of other players' exploration signals to drive the system toward a ``deceptive" Nash equilibrium, while maintaining the stability of the closed-loop system. To extend this insight beyond the duopoly case, in this paper we conduct a comprehensive study of deception mechanisms in N-player oligopoly markets. By leveraging the structure of these games and employing stability techniques for nonlinear dynamical systems, we provide game-theoretic insights into deception and derive specialized results, including stability conditions. These results allow players to systematically adjust their NES dynamics by tuning gains and signal amplitudes, all while ensuring closed-loop stability. Additionally, we introduce novel sufficient conditions to demonstrate that the (practically) stable equilibrium point of the deceptive dynamics corresponds to a true Nash equilibrium of a different game, which we term the ``deceptive game." Our results show that, under the proposed adaptive dynamics with deception, a victim firm may develop a distorted perception of its competitors' product appeal, which could lead to setting suboptimal prices.

In multi-agent autonomous systems, deception is a fundamental concept which characterizes the exploitation of unbalanced information to mislead victims into choosing oblivious actions. This effectively alters the system's long term behavior, leading to outcomes that may be beneficial to the deceiver but detrimental to victim. We study this phenomenon for a class of model-free Nash equilibrium seeking (NES) where players implement independent stochastic exploration signals to learn the pseudogradient flow. In particular, we show that deceptive players who obtain real-time measurements of other players' stochastic perturbation can incorporate this information into their own NES action update, consequentially steering the overall dynamics to a new operating point that could potentially improve the payoffs of the deceptive players. We consider games with quadratic payoff functions, as this restriction allows us to derive a more explicit formulation of the capabilities of the deceptive players. By leveraging results on multi-input stochastic averaging for dynamical systems, we establish local exponential (in probability) convergence for the proposed deceptive NES dynamics. To illustrate our results, we apply them to a two player quadratic game.

We study prescribed-time extremum seeking (PT-ES) for scalar maps in the presence of time delays. The PT-ES problem has been studied by Yilmaz and Krstic in 2023 using chirpy probing and time-varying gains that grow unbounded. To alleviate the gain singularity, in this paper we present an alternative approach, employing delays with bounded time-periodic gains, for achieving prescribed-time convergence to the extremum. Our results are not extensions or refinements of earlier works, but a new methodological direction --applicable even when the map has no delay. The main PT-ES algorithm compensates the map's delay and uses perturbation-based and the Newton (rather than gradient) approaches. With the help of averaging theorems in infinite dimension, specifically Retarded Functional Differential Equations (RFDEs), we conduct a prescribed-time convergence analysis on a suitable averaged target ES system, which contains the time-periodic gains of the map and feedback delays. We further extend our method to multivariable static maps and illustrate our results through numerical simulations.

We study the instability properties of Nesterov's ODE in non-conservative settings, where the driving term is not necessarily the gradient of a potential function. While convergence properties under Nesterov's ODE are well-characterized for optimization settings with gradient-based driving terms, we show that the presence of arbitrarily small non-conservative terms can lead to instability, a phenomenon previously observed empirically via numerical studies in optimization and game-theoretic problems. Our instability analysis combines multi-time scale techniques, such as averaging via variations-of-constants formula, and Floquet Theory, focusing on systems where the vector field is linear and its Helmholtz decomposition reveals a non-vanishing non-conservative component. To resolve the instability issue, the dynamics under non-vanishing non-conservative components, we study a regularization mechanism based on restarting. The resulting system is a hybrid dynamical system that mirrors Nesterov's ODE during intervals of flow, and implements resets of the momentum state through discrete periodic jumps. For this hybrid system, we establish novel explicit bounds on the resetting period that ensure the decrease of a suitable Lyapunov function, guaranteeing not only stability but also "accelerated" convergence rates under suitable smoothness and strong monotonicity properties on the driving term. Numerical simulations support our theoretical results.

The stability of dynamical systems with oscillatory behaviors and well-defined average vector fields has traditionally been studied using averaging theory. These tools have also been applied to hybrid dynamical systems, which combine continuous and discrete dynamics. However, most averaging results for hybrid systems are limited to first-order methods, hindering their use in systems and algorithms that require high-order averaging techniques, such as hybrid Lie-bracket-based extremum seeking algorithms and hybrid vibrational controllers. To address this limitation, we introduce a novel high-order averaging theorem for analyzing the stability of hybrid dynamical systems with high-frequency periodic flow maps. These systems incorporate set-valued flow maps and jump maps, effectively modeling well-posed differential and difference inclusions. By imposing appropriate regularity conditions, we establish results on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$(T,\varepsilon)$</tex-math></inline-formula>-closeness of solutions and semi-global practical asymptotic stability for sets. These theoretical results are then applied to the study of three distinct applications in the context of hybrid model-free control and optimization via Lie-bracket averaging.

We study singularly perturbed systems that exhibit input-to-state stability (ISS) with fixed-time properties in the presence of bounded disturbances. In these systems, solutions converge to the origin within a time frame independent of initial conditions when undisturbed, and to a vicinity of the origin when subjected to bounded disturbances. First, we extend the traditional composite Lyapunov method, commonly applied in singular perturbation theory to analyze asymptotic stability, to include fixed-time ISS. We demonstrate that if both the reduced system and the boundary layer system exhibit fixed-time ISS, and if certain interconnection conditions are met, the entire multi-time scale system retains this fixed-time ISS characteristic, provided the separation of time scales is sufficiently pronounced. Next, we illustrate our findings via analytical and numerical examples, including a novel application in fixed-time feedback optimization for dynamic plants with slowly varying cost functions.

In this paper, we study a novel fixed-time extremum-seeking algorithm that eliminates the need for filters to obtain an appropriate estimation of the gradient of a static map for optimization problems where the cost function is available only via measurements or evaluations. Previous research leveraged these filters to facilitate the application of averaging theory in analyzing the stability properties of the system. Specifically, they were employed to separate, using multi-time scale techniques, the non-smooth terms of the algorithms from the rapidly fluctuating oscillatory terms associated with periodic dithers. This separation was achieved through a singular perturbation argument, where the filter acted as boundary layer system with a sufficiently fast transient. However, since in many practical applications such transient cannot be made arbitrarily fast, and since classic extremum-seeking algorithms are also known to be stable even in the absence of filters, it is natural to ask whether the fixed-time extremum-seeking dynamics can also be simplified by removing the filters while achieving semi-global practical fixed-time convergence properties. This paper addresses this question for scalar quadratic cost functions, providing positive and negative answers depending on the structure of the cost. Additionally, we demonstrate that removing the filters results in average dynamics distinct from the conventional fixed-time gradient flow dynamics found in existing literature. Furthermore, we provide numerical examples to illustrate our findings.

This paper presents a framework that enables analytically shaping the transient behavior of nonlinear dynamical systems, including those with hybrid dynamics combining continuous-time and discrete-time evolution. Our results hinge on the interconnection of the original system with an exogenous dynamic gain system designed to induce a continuous-time deformation of hybrid time domains. Our approach provides conditions that ensure the original system’s stability properties without the dynamic gain are transferable under the continuous-time deformation to the full interconnected dynamics. We develop these results by leveraging tools from hybrid dynamical systems theory, and formulating an appropriate bijective map that relates the solution sets between the original and interconnected systems. To illustrate the approach, we present applications to gradient flow systems and momentum-based optimization techniques with resets, leveraging the framework to customize convergence rates for strictly convex objective functions.

This paper studies the stability and convergence properties of a class of multi-agent concurrent learning (CL) algorithms with momentum and restart. Such algorithms can be integrated as part of the estimation pipelines of data-enabled multi-agent control systems to enhance transient performance while maintaining stability guarantees. However, characterizing restarting policies that yield stable behaviors in decentralized CL systems, especially when the network topology of the communication graph is directed, has remained an open problem. In this paper, we provide an answer to this problem by synergistically leveraging tools from graph theory and hybrid dynamical systems theory. Specifically, we show that under a cooperative richness condition on the overall multi-agent system's data, and by employing coordinated periodic restart with a frequency that is tempered by the level of asymmetry of the communication graph, the resulting decentralized dynamics exhibit robust asymptotic stability properties, characterized in terms of input-to-state stability bounds, and also achieve a desirable transient performance. To demonstrate the practical implications of the theoretical findings, three applications are also presented: cooperative parameter estimation over networks with private data sets, cooperative model-reference adaptive control, and cooperative data-enabled feedback optimization of nonlinear plants.

We consider the problem of achieving prescribed-time stability (PT-S) in a class of hybrid dynamical systems that incorporate switching nonlinear dynamics, exogenous inputs, and resets. By “prescribed-time stability”, we refer to the property of having the main state of the system converge to a particular compact set of interest before a given time defined a priori by the user. We focus on hybrid systems that achieve this property via time-varying gains. For continuous-time systems, this approach has received significant attention in recent years, with various applications in control, optimization, and estimation problems. However, its extensions beyond continuous-time systems have been limited. This gap motivates this paper, which introduces a novel class of switching conditions for switching systems with resets that incorporate time-varying gains, ensuring the PT-S property even in the presence of unstable modes. The analysis leverages tools from hybrid dynamical system’s theory, and a contraction–dilation property that is established for the hybrid time domains of the solutions of the system. We present the model and main results in a general framework, and subsequently apply them to two different problems: (a) PT control of dynamic plants with uncertainty and intermittent feedback; and (b) PT decision-making in non-cooperative switching games using algorithms that incorporate momentum, resets, and dynamic gains. Numerical results are presented to illustrate all our results.

Fixed-time stable dynamical systems are capable of achieving exact convergence to an equilibrium point within a fixed time that is independent of the initial conditions of the system. This property makes them highly appealing for designing control, estimation, and optimization algorithms in applications with stringent performance requirements. However, the set of tools available for analyzing the interconnection of fixed-time stable systems is rather limited compared to their asymptotic counterparts. In this paper, we address some of these limitations by exploiting the emergence of multiple time scales in nonlinear singularly perturbed dynamical systems, where the fast dynamics and the slow dynamics are fixed-time stable on their own. By extending the so-called composite Lyapunov method from asymptotic stability to the context of fixed-time stability, we provide a novel class of Lyapunov-based sufficient conditions to certify fixed-time stability in a class of singularly perturbed dynamical systems. The results are illustrated, analytically and numerically, using a fixed-time gradient flow system interconnected with a fixed-time plant and an additional high-order example.

Stability results for extremum seeking control in $\mathbb{R}^n$ have predominantly been restricted to local or, at best, semi-global practical stability. Extending semi-global stability results of extremum-seeking systems to unbounded sets of initial conditions often demands a stringent global Lipschitz condition on the cost function, which is rarely satisfied by practical applications. In this paper, we address this challenge by leveraging tools from higher-order averaging theory. In particular, we establish a novel second-order averaging result with \emph{global} (practical) stability implications. By leveraging this result, we characterize sufficient conditions on cost functions under which uniform global practical asymptotic stability can be established for a class of extremum-seeking systems acting on static maps. Our sufficient conditions include the case when the gradient of the cost function, rather than the cost function itself, satisfies a global Lipschitz condition, which covers quadratic cost functions. Our results are also applicable to vector fields that are not necessarily Lipschitz continuous at the origin, opening the door to non-smooth Lie-bracket ES dynamics. We illustrate all our results via different analytical and/or numerical examples.

Fixed-time stable dynamical systems are capable of achieving exact convergence to an equilibrium point within a fixed time that is independent of the initial conditions of the system. This property makes them highly appealing for designing control, estimation, and optimization algorithms in applications with stringent performance requirements. How-ever, the set of tools available for analyzing the interconnection of fixed-time stable systems is rather limited compared to their asymptotic counterparts. In this paper, we address some of these limitations by exploiting the emergence of multiple time scales in nonlinear singularly perturbed dynamical systems, where the fast dynamics and the slow dynamics are fixed-time stable on their own. By extending the so-called composite Lyapunov method from asymptotic stability to the context of fixed-time stability, we provide a novel class of Lyapunov-based sufficient conditions to certify fixed-time stability in a class of singularly perturbed dynamical systems. The results are illustrated, analytically and numerically, using a fixed-time gradient flow system interconnected with a fixed-time plant and an additional high-order example.

The convergence of physical and digital systems in modern engineering applications has inevitably led to closed-loop systems that exhibit both continuous-time and discrete-time dynamics. These closed-loop architectures are modeled as hybrid dynamical systems, prevalent across various technological domains, including robotics, power grids, transportation networks, and manufacturing systems. Unlike traditional “smooth” ordinary differential equations or discrete-time recursions, solutions to hybrid dynamical systems are generally discontinuous, lack uniqueness, and have convergence and stability properties that are defined with respect to complex sets. Therefore, effectively designing and controlling such systems, especially under disturbances and uncertainty, is crucial for the development of autonomous and efficient data-driven engineering systems capable of achieving adaptive and self-optimizing behaviors. In this talk, I will delve into recent advancements in the analysis and design of feedback controllers that can achieve such properties in complex scenarios via the synergistic use of adaptive “seeking” dynamics, robust hybrid control, and decision-making algorithms. These controllers can be systematically designed and analyzed using modern tools from hybrid dynamical systems theory, which facilitate the incorporation of “exploration” and “exploitation” behaviors within complex closed-loop systems via multi-time scale tools and perturbation theory. The proposed methodology leads to a family of provably stable and robust algorithms suitable for solving model-free feedback stabilization and decision-making problems in single-agent and multi-agent systems for which smooth feedback solutions fall short.

In socio-technical multi-agent systems, deception exploits privileged information to induce false beliefs in "victims," keeping them oblivious and leading to outcomes detrimental to them or advantageous to the deceiver. We consider model-free Nash-equilibrium-seeking for non-cooperative games with asymmetric information and introduce model-free deceptive algorithms with stability guarantees. In the simplest algorithm, the deceiver includes in his action policy the victim's exploration signal, with an amplitude tuned by an integrator of the regulation error between the deceiver's actual and desired payoff. The integral feedback drives the deceiver's payoff to the payoff's reference value, while the victim is led to adopt a suboptimal action, at which the pseudogradient of the deceiver's payoff is zero. The deceiver's and victim's actions turn out to constitute a "deceptive" Nash equilibrium of a different game, whose structure is managed - in real time - by the deceiver. We examine quadratic, aggregative, and more general games and provide conditions for a successful deception, mutual and benevolent deception, and immunity to deception. Stability results are established using techniques based on averaging and singular perturbations. Among the examples in the paper is a microeconomic duopoly in which the deceiver induces in the victim a belief that the buyers disfavor the deceiver more than they actually do, leading the victim to increase the price above the Nash price, and resulting in an increased profit for the deceiver and a decreased profit for the victim. A study of the deceiver's integral feedback for the desired profit reveals that, in duopolies with equal marginal costs, a deceiver that is greedy for very high profit can attain any such profit, and pursue this with arbitrarily high integral gain (impatiently), irrespective of the market preference for the victim.

We consider the frequency regulation problem for a Virtual Power Plant (VPP) consisting of inverter-interfaced distributed energy resources connected to a power grid, modeled macroscopically, by a conventional generator connected to multiple time-varying loads. To improve the transient performance (settling time, overshoot, etc.) of the frequency response under load disturbances, we introduce a novel Droop Reset Integral Control (DRIC) law that synergistically combines resetting integrators with integral droop controllers (also referred to as proportional integral (PI) control in the literature). We prove the stability of the proposed control scheme, and its robustness to external disturbances, using conditions based on linear matrix inequalities (LMI) that can be numerically verified a priori. Furthermore, we validate the proposed approach using both learned voltage source inverter dynamics and a high-fidelity Simscape model developed by Sandia National Laboratories. Our results show that the DRIC algorithm is able to significantly reduce overshoot, induce zero steady-state error, and decrease settling times up to 7 times that of standard droop and PI control. We also provide heuristic tuning guidelines for the proposed controller, which can be particularly useful for system operators whenever a detailed model of the virtual power plant is unavailable.

—We propose a control scheme via a non-cooperative linear quadratic differential game to coordinate the inverter dynamics of Distributed Energy Resources (DERs) in a microgrid (MG). The MG can provide regulation services in support to the upper-level grid, in addition to serving its own load. The control scheme is designed for the MG to track a power reference, while each DER seeks to minimize its individual cost function subject to learned inverter dynamics and load perturbations. We use a nonlinear high-fidelity model developed by Sandia National Laboratories to learn inverter dynamics. We determine a Nash strategy for the DERs that uses state estimation of a Loop Transfer Recovery. Results show that the control scheme enables savings up to 9.3 to 208 times in the DERs objective cost functions and a time-domain response with no oscillations with up to 3 times faster settling times relative to using droop and PI control.

This paper studies the stability and convergence properties of a class of multi-agent concurrent learning (CL) algorithms with momentum and restart. Such algorithms can be integrated as part of the estimation pipelines of data-enabled multi-agent control systems to enhance transient performance while maintaining stability guarantees. However, characterizing restarting policies that yield stable behaviors in decentralized CL systems, especially when the network topology of the communication graph is directed, has remained an open problem. In this paper, we provide an answer to this problem by synergistically leveraging tools from graph theory and hybrid dynamical systems theory. Specifically, we show that under a cooperative richness condition on the overall multi-agent system's data, and by employing coordinated periodic restart with a frequency that is tempered by the level of asymmetry of the communication graph, the resulting decentralized dynamics exhibit robust asymptotic stability properties, characterized in terms of input-to-state stability bounds, and also achieve a desirable transient performance. To demonstrate the practical implications of the theoretical findings, three applications are also presented: cooperative parameter estimation over networks with private data sets, cooperative model-reference adaptive control, and cooperative data-enabled feedback optimization of nonlinear plants.

Stability results for extremum seeking control in $\mathbb{R}^n$ have predominantly been restricted to local or, at best, semi-global practical stability. Extending semi-global stability results of extremum-seeking systems to unbounded sets of initial conditions often demands a stringent global Lipschitz condition on the cost function, which is rarely satisfied by practical applications. In this paper, we address this challenge by leveraging tools from higher-order averaging theory. In particular, we establish a novel second-order averaging result with \emph{global} (practical) stability implications. By leveraging this result, we characterize sufficient conditions on cost functions under which uniform global practical asymptotic stability can be established for a class of extremum-seeking systems acting on static maps. Our sufficient conditions include the case when the gradient of the cost function, rather than the cost function itself, satisfies a global Lipschitz condition, which covers quadratic cost functions. Our results are also applicable to vector fields that are not necessarily Lipschitz continuous at the origin, opening the door to non-smooth Lie-bracket ES dynamics. We illustrate all our results via different analytical and/or numerical examples.

We introduce a novel concurrent learning (CL) algorithm designed to solve parameter estimation problems within a user-prescribed time frame and by utilizing alternating datasets during the learning process. The algorithm can tackle applications involving switching data sets (including data sets that are completely uninformative) that are updated in real-time as the algorithm operates. To achieve parameter estimation within a specified time independent of the dataset’s richness, the switching algorithm employs dynamic gains. The main result establishes uniform global exponential ultimate boundedness, with an ultimate bound that shrinks to zero as the magnitude of the measurement disturbances decreases. The stability analysis leverages tools from hybrid dynamical systems theory, along with a recently introduced dilation/contraction argument on the hybrid time domains of the solutions. The algorithm and main results are illustrated via a numerical example.

We present stability and recurrence results for a class of stochastic hybrid dynamical systems with oscillating flow maps. These results are developed by introducing averaging tools that parallel those already existing for ordinary differential equations and deterministic hybrid dynamical systems. Such tools can be used to examine the stability properties of the original dynamics based on the properties of a simpler dynamical system constructed from the average of the original oscillating vector field. In this work, we focus on a class of systems for which global stability and recurrence results are achievable under suitable smoothness assumptions on the dynamics. By studying the average stochastic hybrid dynamics using Lyapunov-Foster functions, we derive similar stability and recurrence results for the original stochastic hybrid system.

This letter establishes a novel relationship between a class of recurrent neural networks and certain evolutionary dynamics that emerge in the context of population games. Specifically, it is shown that the output of a recurrent neural network, in the context of classification problems, coincides with the evolution of the population state in a population game. This connection is established with dynamic payoffs and under replicator evolutionary dynamics. The connection provides insights into the neural network's behavior from both dynamical systems and game-theoretical perspectives, aligning with recent literature that suggests that neural network outputs may resemble the Nash equilibria of suitable games. It also uncovers potential connections between the neural network classification problem and mechanism design. To illustrate our results, we present different numerical experiments in the context of classification problems.

We present stability and recurrence results for a class of stochastic hybrid dynamical systems with oscillating flow maps. These results are developed by introducing averaging tools that parallel those already existing for ordinary differential equations and deterministic hybrid dynamical systems. Such tools can be used to examine the stability properties of the original dynamics based on the properties of a simpler dynamical system constructed from the average of the original oscillating vector field. In this work, we focus on a class of systems for which global stability and recurrence results are achievable under suitable smoothness assumptions on the dynamics. By studying the average stochastic hybrid dynamics using Lyapunov-Foster functions, we derive similar stability and recurrence results for the original stochastic hybrid system.

Broadly speaking, his research interests span the analysis and design of adaptive, robust, and highperformance feedback-based dynamic mechanisms for control, optimization, estimation, and learning, with an emphasis on hybrid dynamical systems. His laboratory works on the development of foundational theories and algorithms for the safe and efficient deployment of feedback-enabled autonomous machines in different engineering, societal, and biological systems. To achieve this overarching goal, his research combines tools from control theory, dynamical systems, optimization, game theory, and network science. Application domains include cyber-physical systems (power

We introduce new sufficient conditions for verifying stability and recurrence properties in singularly perturbed stochastic hybrid dynamical systems. Specifically, we focus on hybrid systems with deterministic continuous-time dynamics that exhibit multiple time scales and are modeled by constrained differential inclusions, as well as discrete-time dynamics modeled by constrained difference inclusions with random inputs. By assuming regularity and causality of the dynamics and their solutions, respectively, we propose a suitable class of composite nonsmooth Lagrange-Foster and Lyapunov-Foster functions that can certify stability and recurrence using simpler functions related to the slow and fast dynamics of the system. We establish the stability properties with respect to compact sets, while the recurrence properties are studied only for open sets.

The stability of dynamical systems with oscillatory behaviors and well-defined average vector fields has traditionally been studied using averaging theory. These tools have also been applied to hybrid dynamical systems, which combine continuous and discrete dynamics. However, most averaging results for hybrid systems are limited to first-order methods, hindering their use in systems and algorithms that require high-order averaging techniques, such as hybrid Lie-bracket-based extremum seeking algorithms and hybrid vibrational controllers. To address this limitation, we introduce a novel high-order averaging theorem for analyzing the stability of hybrid dynamical systems with high-frequency periodic flow maps. These systems incorporate set-valued flow maps and jump maps, effectively modeling well-posed differential and difference inclusions. By imposing appropriate regularity conditions, we establish results on $(T,\varepsilon)$-closeness of solutions and semi-global practical asymptotic stability for sets. These theoretical results are then applied to the study of three distinct applications in the context of hybrid model-free control and optimization via Lie-bracket averaging.

This paper proves some rigidity results for shrinking and expanding Ricci solitons. First, we demonstrate that compact shrinking Ricci solitons are Einstein if we control the maximum value of the potential function. Then, we prove some rigidity results for non-compact gradient expanding and shrinking Ricci solitons with pinched Ricci (or scalar) curvature, assuming an asymptotic condition on the scalar curvature at infinity.

We introduce a class of decentralized high-order pricing dynamics (HOPD) for the solution of optimal incentive problems in affine congestion games with full resource utilization. The dynamics incorporate momentum and decentralized coordinated resets to achieve better transient performance compared to traditional first-order gradient-based pricing algorithms. The proposed dynamics are studied using tools from graph theory, game theory, and hybrid dynamical systems theory. Our main results establish suitable stability and convergence properties with respect to the set of incentives that generate Nash flows that also maximize the social welfare function of the game. The theoretical results are illustrated via numerical examples in two different types of communication graphs, highlighting the effect of the communication topology and the coordination between players on the transient performance of the HOPD.

This paper introduces a class of model-free feedback methods for solving generic constrained optimization problems where the specific mathematical forms of the objective and constraint functions are not available. The proposed methods, termed Projected Zeroth-Order (P-ZO) dynamics, incorporate projection maps into a class of continuous-time model-free dynamics that make use of periodic dithering for the purpose of gradient learning. In particular, the proposed P-ZO algorithms can be interpreted as new extremum-seeking algorithms that autonomously drive an unknown system toward a neighborhood of the set of solutions of an optimization problem using only output feedback, while systematically guaranteeing that the input trajectories remain in a feasible set for all times. In this way, the P-ZO algorithms can properly handle hard and asymptotical constraints in model-free optimization problems without using penalty terms or barrier functions. Moreover, the proposed dynamics have suitable robustness properties with respect to small bounded additive disturbances on the states and dynamics, a property that is fundamental for practical real-world implementations. Additional tracking results for time-varying and switching cost functions are also derived under stronger convexity and smoothness assumptions and using tools from hybrid dynamical systems. Numerical examples are presented throughout the paper to illustrate the above results.

Multi-time scale techniques based on singular perturbations and averaging theory are among the most powerful tools developed for the synthesis and analysis of feedback control algorithms. This paper introduces some of the recent advances in singular perturbation theory and averaging theory for continuous-time dynamical systems modeled as ordinary differential equations (ODEs), as well as for hybrid dynamical systems that combine continuous-time dynamics and discrete-time dynamics. Novel multi-time scale analytical tools based on higher-order averaging and singular perturbation theory are also discussed and illustrated via different examples. In the context of hybrid dynamical systems, a class of sufficient Lyapunov-based conditions for global stability results are also presented. The analytical tools are illustrated through various new architectures and algorithms within the context of adaptive and extremum-seeking systems. These tools are suitable for the study of model-free optimization and stabilization problems that require the synergistic use of continuous-time and discrete-time feedback. The paper aims to acquaint the reader with a range of modern tools for studying multi-time scale phenomena in optimization and control systems, providing some guidelines for future research in this field.