arXiv subjects

Sandro Zampieri

Publications and source records attributed to Sandro Zampieri.

At least 19 recordsLinked to original sources

arXiv2026

A Dynamical Theory of Sequential Retrieval in Input-Driven Hopfield Networks

Reasoning is the ability to integrate internal states and external inputs in a meaningful and semantically consistent flow. Contemporary machine learning (ML) systems increasingly rely on such sequential reasoning, from language understanding to multi-modal generation, often operating over dictionaries of prototypical patterns reminiscent of associative memory models. Understanding retrieval and sequentiality in associative memory models provides a powerful bridge to gain insight into ML reasoning. While the static retrieval properties of associative memory models are well understood, the theoretical foundations of sequential retrieval and multi-memory integration remain limited, with existing studies largely relying on numerical evidence. This work develops a dynamical theory of sequential reasoning in Hopfield networks. We consider the recently proposed input-driven plasticity (IDP) Hopfield network and analyze a two-timescale architecture coupling fast associative retrieval with slow reasoning dynamics. We derive explicit conditions for self-sustained memory transitions, including gain thresholds, escape times, and collapse regimes. Together, these results provide a principled mathematical account of sequentiality in associative memory models, bridging classical Hopfield dynamics and modern reasoning architectures.

cs.NE↗

arXiv2025

Geometric SSM: LTI State Space Models for Selective Tasks

A key claim in recent work on Selective State Space Models is that selectivity, the ability to focus on relevant information while filtering irrelevant inputs, requires breaking the Linear Time-Invariant (LTI) property through time-varying dynamics. We challenge this claim by demonstrating that LTI systems can achieve selectivity when designed using principles from geometric control. We introduce the Geometric SSM, in which different input patterns excite distinct invariant subspaces of the dynamics. Unlike Mamba's memoryless selection mechanism, our approach employs a dynamic residual generator that maintains temporal memory, enabling recognition of multi-token patterns without time-varying system matrices. The Geometric SSM achieves near-perfect performance on a novel extended induction head task where Mamba fails, while preserving efficient FFT-based training. Our results demonstrate that geometric control theory can inform the design of novel selective sequence models that combine theoretical rigor with practical efficiency.

eess.SY↗

arXiv2025

Anti-windup design for internal model online constrained optimization

This paper proposes a novel algorithmic design procedure for online constrained optimization grounded in control-theoretic principles. By integrating the Internal Model Principle (IMP) with an anti-windup compensation mechanism, the proposed Projected-Internal Model Anti-Windup (P-IMAW) gradient descent exploits a partial knowledge of the temporal evolution of the cost function to enhance tracking performance. The algorithm is developed through a structured synthesis procedure: first, a robust controller leveraging the IMP ensures asymptotic convergence in the unconstrained setting. Second, an anti-windup augmentation guarantees stability and performance in the presence of the projection operator needed to satisfy the constraints. The effectiveness of the proposed approach is demonstrated through numerical simulations comparing it against other classical techniques.

math.OC↗

arXiv2025

Controllable Neural Architectures for Multi-Task Control

This paper studies a multi-task control problem where multiple linear systems are to be regulated by a single non-linear controller. In particular, motivated by recent advances in multi-task learning and the design of brain-inspired architectures, we consider a neural controller with (smooth) ReLU activation function. The parameters of the controller are a connectivity matrix and a bias vector: although both parameters can be designed, the connectivity matrix is constant while the bias vector can be varied and is used to adapt the controller across different control tasks. The bias vector determines the equilibrium of the neural controller and, consequently, of its linearized dynamics. Our multi-task control strategy consists of designing the connectivity matrix and a set of bias vectors in a way that the linearized dynamics of the neural controller for the different bias vectors provide a good approximation of a set of desired controllers. We show that, by properly choosing the bias vector, the linearized dynamics of the neural controller can replicate the dynamics of any single, linear controller. Further, we design gradient-based algorithms to train the parameters of the neural controller, and we provide upper and lower bounds for the performance of our neural controller. Finally, we validate our results using different numerical examples.

eess.SY↗

arXiv2024

Stochastic models for online optimization

In this paper, we propose control-theoretic methods as tools for the design of online optimization algorithms that are able to address dynamic, noisy, and partially uncertain time-varying quadratic objective functions. Our approach introduces two algorithms specifically tailored for scenarios where the cost function follows a stochastic linear model. The first algorithm is based on a Kalman filter-inspired approach, leveraging state estimation techniques to account for the presence of noise in the evolution of the objective function. The second algorithm applies $\mathcal{H}_\infty$-robust control strategies to enhance performance under uncertainty, particularly in cases in which model parameters are characterized by a high variability. Through numerical experiments, we demonstrate that our algorithms offer significant performance advantages over the traditional gradient-based method and also over the optimization strategy proposed in arXiv:2205.13932 based on deterministic models.

math.OC↗

arXiv2024

Firing Rate Models as Associative Memory: Excitatory-Inhibitory Balance for Robust Retrieval

Firing rate models are dynamical systems widely used in applied and theoretical neuroscience to describe local cortical dynamics in neuronal populations. By providing a macroscopic perspective of neuronal activity, these models are essential for investigating oscillatory phenomena, chaotic behavior, and associative memory processes. Despite their widespread use, the application of firing rate models to associative memory networks has received limited mathematical exploration, and most existing studies are focused on specific models. Conversely, well-established associative memory designs, such as Hopfield networks, lack key biologically-relevant features intrinsic to firing rate models, including positivity and interpretable synaptic matrices that reflect excitatory and inhibitory interactions. To address this gap, we propose a general framework that ensures the emergence of re-scaled memory patterns as stable equilibria in the firing rate dynamics. Furthermore, we analyze the conditions under which the memories are locally and globally asymptotically stable, providing insights into constructing biologically-plausible and robust systems for associative memory retrieval.

q-bio.NC↗

arXiv2024

Input-Driven Dynamics for Robust Memory Retrieval in Hopfield Networks

The Hopfield model provides a mathematically idealized yet insightful framework for understanding the mechanisms of memory storage and retrieval in the human brain. This model has inspired four decades of extensive research on learning and retrieval dynamics, capacity estimates, and sequential transitions among memories. Notably, the role and impact of external inputs has been largely underexplored, from their effects on neural dynamics to how they facilitate effective memory retrieval. To bridge this gap, we propose a novel dynamical system framework in which the external input directly influences the neural synapses and shapes the energy landscape of the Hopfield model. This plasticity-based mechanism provides a clear energetic interpretation of the memory retrieval process and proves effective at correctly classifying highly mixed inputs. Furthermore, we integrate this model within the framework of modern Hopfield architectures, using this connection to elucidate how current and past information are combined during the retrieval process. Finally, we embed both the classic and the new model in an environment disrupted by noise and compare their robustness during memory retrieval.

q-bio.NC↗

arXiv2024

An invariance principle based concentration result for large-scale stochastic pairwise interaction network systems

We study stochastic pairwise interaction network systems whereby a finite population of agents, identified with the nodes of a graph, update their states in response to both individual mutations and pairwise interactions with their neighbors. The considered class of systems include the main epidemic models -such as the SIS, SIR, and SIRS models-, certain social dynamics models -such as the voter and anti-voter models-, as well as evolutionary dynamics on graphs. Since these stochastic systems fall into the class of finite-state Markov chains, they always admit stationary distributions. We analyze the asymptotic behavior of these stationary distributions in the limit as the population size grows large while the interaction network maintains certain mixing properties. Our approach relies on the use of Lyapunov-type functions to obtain concentration results on these stationary distributions. Notably, our results are not limited to fully mixed population models, as they do apply to a much broader spectrum of interaction network structures, including, e.g., Erd\"oos-R\'enyi random graphs.

eess.SY↗

arXiv2023

Dynamic Brain Networks with Prescribed Functional Connectivity

In this paper, we consider stable stochastic linear systems modeling whole-brain resting-state dynamics. We parametrize the state matrix of the system (effective connectivity) in terms of its steady-state covariance matrix (functional connectivity) and a skew-symmetric matrix $S$. We examine how the matrix $S$ influences some relevant dynamic properties of the system. Specifically, we show that a large $S$ enhances the degree of stability and excitability of the system, and makes the latter more responsive to high-frequency inputs.

eess.SY↗

arXiv2023

A Control Theoretical Approach to Online Constrained Optimization

In this paper we focus on the solution of online problems with time-varying, linear equality and inequality constraints. Our approach is to design a novel online algorithm by leveraging the tools of control theory. In particular, for the case of equality constraints only, using robust control we design an online algorithm with asymptotic convergence to the optimal trajectory, differently from the alternatives that achieve non-zero tracking error. When also inequality constraints are present, we show how to modify the proposed algorithm to account for the wind-up induced by the nonnegativity constraints on the dual variables. We report numerical results that corroborate the theoretical analysis, and show how the proposed approach outperforms state-of-the-art algorithms both with equality and inequality constraints.

math.OC↗

arXiv2022

Internal Model-Based Online Optimization

In this paper we propose a model-based approach to the design of online optimization algorithms, with the goal of improving the tracking of the solution trajectory (trajectories) w.r.t. state-of-the-art methods. We focus first on quadratic problems with a time-varying linear term, and use digital control tools (a robust internal model principle) to propose a novel online algorithm that can achieve zero tracking error by modeling the cost with a dynamical system. We prove the convergence of the algorithm for both strongly convex and convex problems. We further discuss the sensitivity of the proposed method to model uncertainties and quantify its performance. We discuss how the proposed algorithm can be applied to general (non-quadratic) problems using an approximate model of the cost, and analyze the convergence leveraging the small gain theorem. We present numerical results that showcase the superior performance of the proposed algorithms over previous methods for both quadratic and non-quadratic problems.

math.OC↗

arXiv2019

On the existence and linear approximation of the power flow solution in power distribution networks

We consider the problem of deriving an explicit approximate solution of the nonlinear power equations that describe a balanced power distribution network. We give sufficient conditions for the existence of a practical solution to the power flow equations, and we propose an approximation that is linear in the active and reactive power demands of the PQ buses. For this approximation, which is valid for generic power line impedances and grid topology, we derive a bound on the approximation error as a function of the grid parameters. We illustrate the quality of the approximation via simulations, we show how it can also model the presence of voltage controlled (PV) buses, and we discuss how it generalizes the DC power flow model to lossy networks.

math.OC↗

arXiv2019

Brain Controllability: not a slam dunk yet

In our recent article (Tu et al., Warnings and caveats in brain controllability, arXiv:1705.08261) we provided quantitative evidence to show that there are warnings and caveats in the way Gu and collaborators (Gu et al. Controllability of structural brain networks. Nature communications 6 (2015): 8414) define brain controllability. The comment by Pasqualetti et al. (Pasqualetti et al. RE: Warnings and Caveats in Brain Controllability. NeuroImage 297 (2019), 586-588) confirms the need to go beyond the methodology and approach presented in Gu et al. original work. In fact, they recognize that the source of confusion is due to the fact that assessing controllability via numerical analysis typically leads to ill-conditioned problems, and thus often generates results that are difficult to interpret. This is indeed the first warning we discussed: our work was not meant to prove that brain networks are not controllable from one node, rather we wished to highlight that the one node controllability framework and all consequent results were not properly justified based on the methodology presented in Gu et al. We used in our work the same method of Gu et al. not because we believe it is the best methodology, but because we extensively investigated it with the aim of replicating, testing and extending their results. And the warning and caveats we have proposed are the results of this investigation.

q-bio.QM↗

arXiv2019

Efficient Communication over Complex Dynamical Networks: The Role of Matrix Non-Normality

In both natural and engineered systems, communication often occurs dynamically over networks ranging from highly structured grids to largely disordered graphs. To use, or comprehend the use of, networks as efficient communication media requires understanding of how they propagate and transform information in the face of noise. Here, we develop a framework that enables us to examine how network structure, noise, and interference between consecutive packets jointly determine transmission performance in networks with linear dynamics at single nodes and arbitrary topologies. Mathematically normal networks, which can be decomposed into separate low-dimensional information channels, suffer greatly from readout and interference noise. Interestingly, most details of their wiring have no impact on transmission quality. Non-normal networks, however, can largely cancel the effect of noise by transiently amplifying select input dimensions while ignoring others, resulting in higher net information throughput. Our theory could inform the design of new communication networks, as well as the optimal use of existing ones.

physics.soc-ph↗

arXiv2017

Warnings and Caveats in Brain Controllability

In this work we challenge the main conclusions of Gu et al work (Controllability of structural brain networks. Nature communications 6, 8414, doi:10.1038/ncomms9414, 2015) on brain controllability. Using the same methods and analyses on four datasets we find that the minimum set of nodes to control brain networks is always larger than one. We also find that the relationships between the average/modal controllability and weighted degrees also hold for randomized data and the there are not specific roles played by Resting State Networks in controlling the brain. In conclusion, we show that there is no evidence that topology plays specific and unique roles in the controllability of brain networks. Accordingly, Gu et al. interpretation of their results, in particular in terms of translational applications (e.g. using single node controllability properties to define target region(s) for neurostimulation) should be revisited. Though theoretically intriguing, our understanding of the relationship between controllability and structural brain network remains elusive.

q-bio.NC↗

arXiv2017

On the Dynamics of Deterministic Epidemic Propagation over Networks

In this work we review a class of deterministic nonlinear models for the propagation of infectious diseases over contact networks with strongly-connected topologies. We consider network models for susceptible-infected (SI), susceptible-infected-susceptible (SIS), and susceptible-infected-recovered (SIR) settings. In each setting, we provide a comprehensive nonlinear analysis of equilibria, stability properties, convergence, monotonicity, positivity, and threshold conditions. For the network SI setting, specific contributions include establishing its equilibria, stability, and positivity properties. For the network SIS setting, we review a well-known deterministic model, provide novel results on the computation and characterization of the endemic state (when the system is above the epidemic threshold), and present alternative proofs for some of its properties. Finally, for the network SIR setting, we propose novel results for transient behavior, threshold conditions, stability properties, and asymptotic convergence. These results are analogous to those well-known for the scalar case. In addition, we provide a novel iterative algorithm to compute the asymptotic state of the network SIR system.

cs.SI↗

arXiv2016

On the Role of Network Centrality in the Controllability of Complex Networks

In recent years complex networks have gained increasing attention in different fields of science and engineering. The problem of controlling these networks is an interesting and challenging problem to investigate. In this paper we look at the controllability problem focusing on the energy needed for the control. Precisely not only we want to analyze whether a network can be controlled, but we also want to establish whether the control can be performed using a limited amount of energy. We restrict our study to irreducible and (marginally) stable networks and we find that the leading right and left eigenvectors of the network matrix play a crucial role in this analysis. Interestingly, our results suggest the existence of a connection between controllability and network centrality, a well-known concept in network science. In case the network is reversible, the latter connection involves the PageRank, an extensively studied type of centrality measure. Finally, the proposed results are applied to examples concerning random graphs.

math.OC↗

arXiv2014

Distributed reactive power feedback control for voltage regulation and loss minimization

We consider the problem of exploiting the microgenerators dispersed in the power distribution network in order to provide distributed reactive power compensation for power losses minimization and voltage regulation. In the proposed strategy, microgenerators are smart agents that can measure their phasorial voltage, share these data with the other agents on a cyber layer, and adjust the amount of reactive power injected into the grid, according to a feedback control law that descends from duality-based methods applied to the optimal reactive power flow problem. Convergence to the configuration of minimum losses and feasible voltages is proved analytically for both a synchronous and an asynchronous version of the algorithm, where agents update their state independently one from the other. Simulations are provided in order to illustrate the performance and the robustness of the algorithm, and the innovative feedback nature of such strategy is discussed.

math.OC↗