-
Optimal distributed control with stability guarantees by training a network of neural closed-loop maps
Authors:
Danilo Saccani,
Leonardo Massai,
Luca Furieri,
Giancarlo Ferrari-Trecate
Abstract:
This paper proposes a novel approach to improve the performance of distributed nonlinear control systems while preserving stability by leveraging Deep Neural Networks (DNNs). We build upon the Neural System Level Synthesis (Neur-SLS) framework and introduce a method to parameterize stabilizing control policies that are distributed across a network topology. A distinctive feature is that we iterati…
▽ More
This paper proposes a novel approach to improve the performance of distributed nonlinear control systems while preserving stability by leveraging Deep Neural Networks (DNNs). We build upon the Neural System Level Synthesis (Neur-SLS) framework and introduce a method to parameterize stabilizing control policies that are distributed across a network topology. A distinctive feature is that we iteratively minimize an arbitrary control cost function through an unconstrained optimization algorithm, all while preserving the stability of the overall network architecture by design. This is achieved through two key steps. First, we establish a method to parameterize interconnected Recurrent Equilibrium Networks (RENs) that guarantees a bounded $\mathcal{L}_2$ gain at the network level. This ensures stability. Second, we demonstrate how the information flow within the network is preserved, enabling a fully distributed implementation where each subsystem only communicates with its neighbors. To showcase the effectiveness of our approach, we present a simulation of a distributed formation control problem for a fleet of vehicles. The simulation demonstrates how the proposed neural controller enables the vehicles to maintain a desired formation while navigating obstacles and avoiding collisions, all while guaranteeing network stability.
△ Less
Submitted 3 April, 2024;
originally announced April 2024.
-
Unconstrained learning of networked nonlinear systems via free parametrization of stable interconnected operators
Authors:
Leonardo Massai,
Danilo Saccani,
Luca Furieri,
Giancarlo Ferrari-Trecate
Abstract:
This paper characterizes a new parametrization of nonlinear networked incrementally $L_2$-bounded operators in discrete time. The distinctive novelty is that our parametrization is \emph{free} -- that is, a sparse large-scale operator with bounded incremental $L_2$ gain is obtained for any choice of the real values of our parameters. This property allows one to freely search over optimal parameter…
▽ More
This paper characterizes a new parametrization of nonlinear networked incrementally $L_2$-bounded operators in discrete time. The distinctive novelty is that our parametrization is \emph{free} -- that is, a sparse large-scale operator with bounded incremental $L_2$ gain is obtained for any choice of the real values of our parameters. This property allows one to freely search over optimal parameters via unconstrained gradient descent, enabling direct applications in large-scale optimal control and system identification. Further, we can embed prior knowledge about the interconnection topology and stability properties of the system directly into the large-scale distributed operator we design. Our approach is extremely general in that it can seamlessly encapsulate and interconnect state-of-the-art Neural Network (NN) parametrizations of stable dynamical systems. To demonstrate the effectiveness of this approach, we provide a simulation example showcasing the identification of a networked nonlinear system. The results underscore the superiority of our free parametrizations over standard NN-based identification methods where a prior over the system topology and local stability properties are not enforced.
△ Less
Submitted 27 March, 2024; v1 submitted 23 November, 2023;
originally announced November 2023.
-
Impact of interface traps on charge noise, mobility and percolation density in Ge/SiGe heterostructures
Authors:
L. Massai,
B. Hetényi,
M. Mergenthaler,
F. J. Schupp,
L. Sommer,
S. Paredes,
S. W. Bedell,
P. Harvey-Collard,
G. Salis,
A. Fuhrer,
N. W. Hendrickx
Abstract:
Hole spins in Ge/SiGe heterostructure quantum dots have emerged as promising qubits for quantum computation. The strong spin-orbit coupling (SOC), characteristic of heavy-hole states in Ge, enables fast and all-electrical qubit control. However, SOC also increases the susceptibility of spin qubits to charge noise. While qubit coherence can be significantly improved by operating at sweet spots with…
▽ More
Hole spins in Ge/SiGe heterostructure quantum dots have emerged as promising qubits for quantum computation. The strong spin-orbit coupling (SOC), characteristic of heavy-hole states in Ge, enables fast and all-electrical qubit control. However, SOC also increases the susceptibility of spin qubits to charge noise. While qubit coherence can be significantly improved by operating at sweet spots with reduced hyperfine or charge noise sensitivity, the latter ultimately limits coherence, underlining the importance of understanding and reducing charge noise at its source. In this work, we study the voltage-induced hysteresis commonly observed in SiGe-based quantum devices and show that the dominant charge fluctuators are localized at the semiconductor-oxide interface. By applying increasingly negative gate voltages to Hall bar and quantum dot devices, we investigate how the hysteretic filling of interface traps impacts transport metrics and charge noise. We find that the gate-induced accumulation and trap** of charge at the SiGe-oxide interface leads to an increased electrostatic disorder, as probed by transport measurements, as well as the activation of low-frequency relaxation dynamics, resulting in slow drifts and increased charge noise levels. Our results highlight the importance of a conservative device tuning strategy and reveal the critical role of the semiconductor-oxide interface in SiGe heterostructures for spin qubit applications.
△ Less
Submitted 9 October, 2023;
originally announced October 2023.
-
Sweet-spot operation of a germanium hole spin qubit with highly anisotropic noise sensitivity
Authors:
N. W. Hendrickx,
L. Massai,
M. Mergenthaler,
F. Schupp,
S. Paredes,
S. W. Bedell,
G. Salis,
A. Fuhrer
Abstract:
Spin qubits defined by valence band hole states comprise an attractive candidate for quantum information processing due to their inherent coupling to electric fields enabling fast and scalable qubit control. In particular, heavy holes in germanium have shown great promise, with recent demonstrations of fast and high-fidelity qubit operations. However, the mechanisms and anisotropies that underlie…
▽ More
Spin qubits defined by valence band hole states comprise an attractive candidate for quantum information processing due to their inherent coupling to electric fields enabling fast and scalable qubit control. In particular, heavy holes in germanium have shown great promise, with recent demonstrations of fast and high-fidelity qubit operations. However, the mechanisms and anisotropies that underlie qubit driving and decoherence are still mostly unclear. Here, we report on the highly anisotropic heavy-hole $g$-tensor and its dependence on electric fields, allowing us to relate both qubit driving and decoherence to an electric modulation of the $g$-tensor. We also confirm the predicted Ising-type hyperfine interaction but show that qubit coherence is ultimately limited by $1/f$ charge noise. Finally, we operate the qubit at low magnetic field and measure a dephasing time of $T_2^*=9.2$ $μ$s, while maintaining a single-qubit gate fidelity of 99.94 %, that remains well above 99 % at an operation temperature T>1 K. This understanding of qubit driving and decoherence mechanisms are key for the design and operation of scalable and highly coherent hole qubit arrays.
△ Less
Submitted 24 November, 2023; v1 submitted 22 May, 2023;
originally announced May 2023.
-
Pure circularly polarized light emission from waveguide microring resonators
Authors:
Leonardo Massai,
Tom Schatteburg,
Jonathan P. Home,
Karan K. Mehta
Abstract:
Circularly polarized light plays a key role in many applications including spectroscopy, microscopy, and control of atomic systems. Particularly in the latter, high polarization purity is often required. Integrated technologies for atomic control are progressing rapidly, but while integrated photonics can generate fields with pure linear polarization, integrated generation of highly pure circular…
▽ More
Circularly polarized light plays a key role in many applications including spectroscopy, microscopy, and control of atomic systems. Particularly in the latter, high polarization purity is often required. Integrated technologies for atomic control are progressing rapidly, but while integrated photonics can generate fields with pure linear polarization, integrated generation of highly pure circular polarization states has not been addressed. Here, we show that waveguide microring resonators, perturbed with azimuthal gratings and thereby emitting beams carrying optical orbital angular momentum, can generate radiated fields of high circular polarization purity. We achieve this in a passive device by taking advantage of symmetries of the structure and radiated modes, and directly utilizing both transverse and longitudinal field components of the guided modes. On the axis of emission and at maximum intensity, we measure an average polarization impurity of $1.0 \times 10^{-3}$ in relative intensity across the resonance FWHM, and observe impurities below $10^{-4}$ in this range. This constitutes a significant improvement over the ${\sim}10^{-2}$ impurity demonstrated in previous work on integrated devices. Photonic structures allowing high circular polarization purity may assist in realizing high-fidelity control and measurement in atomic quantum systems.
△ Less
Submitted 6 July, 2022;
originally announced July 2022.
-
Equilibria in Network Constrained Energy Markets
Authors:
Leonardo Massai,
Giacomo Como,
Fabio Fagnani
Abstract:
We study an energy market composed of producers who compete to supply energy to different markets and want to maximize their profits. The energy market is modeled by a graph representing a constrained power network where nodes represent the markets and links are the physical lines with a finite capacity connecting them. Producers play a networked Cournot game on such a network together with a cent…
▽ More
We study an energy market composed of producers who compete to supply energy to different markets and want to maximize their profits. The energy market is modeled by a graph representing a constrained power network where nodes represent the markets and links are the physical lines with a finite capacity connecting them. Producers play a networked Cournot game on such a network together with a centralized authority, called market maker, that facilitates the trade between geographically separate markets via the constrained power network and aims to maximize a certain welfare function. We first prove a general result that links the optimal action of the market maker with the capacity constraint enforced on the power network. Under mild assumptions, we study the existence and uniqueness of Nash equilibria and exploit our general result to prove a connection between capacity bottlenecks in the power network and the emergence of price differences between different markets that are separated by saturated lines, a phenomenon that is often observed in real power networks.
△ Less
Submitted 30 November, 2022; v1 submitted 15 June, 2022;
originally announced June 2022.
-
Evaluation of semantic relations impact in query expansion-based retrieval systems
Authors:
Lorenzo Massai
Abstract:
With the increasing demand of intelligent systems capable of operating in different contexts (e.g. users on the move) the correct interpretation of the user-need by such systems has become crucial to give consistent answers to the user questions. The most effective applications addressing such task are in the fields of natural language processing and semantic expansion of terms. These techniques a…
▽ More
With the increasing demand of intelligent systems capable of operating in different contexts (e.g. users on the move) the correct interpretation of the user-need by such systems has become crucial to give consistent answers to the user questions. The most effective applications addressing such task are in the fields of natural language processing and semantic expansion of terms. These techniques are aimed at estimating the goal of an input query reformulating it as an intent, commonly relying on textual resources built exploiting different semantic relations like \emph{synonymy}, \emph{antonymy} and many others. The aim of this paper is to generate such resources using the labels of a given taxonomy as source of information. The obtained resources are integrated into a plain classifier for reformulating a set of input queries as intents and tracking the effect of each relation, in order to quantify the impact of each semantic relation on the classification. As an extension to this, the best tradeoff between improvement and noise introduction when combining such relations is evaluated. The assessment is made generating the resources and their combinations and using them for tuning the classifier which is used to reformulate the user questions as labels. The evaluation employs a wide and varied taxonomy as a use-case, exploiting its labels as basis for the semantic expansion and producing several corpora with the purpose of enhancing the pseudo-queries estimation.
△ Less
Submitted 15 December, 2023; v1 submitted 30 March, 2022;
originally announced March 2022.
-
Equilibria and Systemic Risk in Saturated Networks
Authors:
Leonardo Massai,
Giacomo Como,
Fabio Fagnani
Abstract:
We undertake a fundamental study of network equilibria modeled as solutions of fixed point equations for monotone linear functions with saturation nonlinearities. The considered model extends one originally proposed to study systemic risk in networks of financial institutions interconnected by mutual obligations and is one of the simplest continuous models accounting for shock propagation phenomen…
▽ More
We undertake a fundamental study of network equilibria modeled as solutions of fixed point equations for monotone linear functions with saturation nonlinearities. The considered model extends one originally proposed to study systemic risk in networks of financial institutions interconnected by mutual obligations and is one of the simplest continuous models accounting for shock propagation phenomena and cascading failure effects. It also characterizes Nash equilibria of constrained quadratic network games with strategic complementarities. We first derive explicit expressions for network equilibria and prove necessary and sufficient conditions for their uniqueness encompassing and generalizing results available in the literature. Then, we study jump discontinuities of the network equilibria when the exogenous flows cross certain regions of measure 0 representable as graphs of continuous functions. Finally, we discuss some implications of our results in the two main motivating applications. In financial networks, this bifurcation phenomenon is responsible for how small shocks in the assets of a few nodes can trigger major aggregate losses to the system and cause the default of several agents. In constrained quadratic network games, it induces a blow-up behavior of the sensitivity of Nash equilibria with respect to the individual benefits.
△ Less
Submitted 18 January, 2021; v1 submitted 10 December, 2019;
originally announced December 2019.
-
Stability and phase transitions of dynamical flow networks with finite capacities
Authors:
Leonardo Massai,
Giacomo Como,
Fabio Fagnani
Abstract:
We study deterministic continuous-time lossy dynamical flow networks with constant exogenous demands, fixed routing, and finite flow and buffer capacities. In the considered model, when the total net flow in a cell ---consisting of the difference between the total flow directed towards it minus the outflow from it--- exceeds a certain capacity constraint, then the exceeding part of it leaks out of…
▽ More
We study deterministic continuous-time lossy dynamical flow networks with constant exogenous demands, fixed routing, and finite flow and buffer capacities. In the considered model, when the total net flow in a cell ---consisting of the difference between the total flow directed towards it minus the outflow from it--- exceeds a certain capacity constraint, then the exceeding part of it leaks out of the system. The ensuing network flow dynamics is a linear saturated system with compact state space that we analyze using tools from monotone systems and contraction theory. Specifically, we prove that there exists a set of equilibria that is globally asymptotically stable. Such equilibrium set reduces to a single globally asymptotically stable equilibrium for generic exogenous demand vectors. Moreover, we show that the critical exogenous demand vectors giving rise to non-unique equilibria correspond to phase transitions in the asymptotic behavior of the dynamical flow network.
△ Less
Submitted 4 December, 2019;
originally announced December 2019.
-
Risk measures and Margining control
Authors:
Giuseppe Carlo Calafiore,
Leonardo Massai
Abstract:
This document constitutes the final report of the contractual activity between Directa SIM and Dipartimento di Automatica e Informatica, Politecnico di Torino, on the research topic titled "quantificazione del rischio di un portafoglio di strumenti finanziari per trading online su device fissi e mobili."
This document constitutes the final report of the contractual activity between Directa SIM and Dipartimento di Automatica e Informatica, Politecnico di Torino, on the research topic titled "quantificazione del rischio di un portafoglio di strumenti finanziari per trading online su device fissi e mobili."
△ Less
Submitted 29 August, 2016;
originally announced August 2016.
