Papers for

control systems engineers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Lean software helps verify math proofs in system identification research

From Pen-and-Paper to Machine-Checked Proofs: Lean for System Identification

Abstract: Mathematical proofs in scientific papers are often long and intricate, relying on chains of references to earlier results. This makes it difficult for readers and reviewers to verify each and every step. Interactive theorem proving offers a complement: the theorem and its proof are written as code, and every inference is checked by a small, trusted piece of software called the kernel. This paper serves as an introduction to the Lean 4 proof assistant for a system identification and control audience, assuming no prior exposure to formal methods. We first show how a Lean proof is read using two elementary examples: the first on identifiability, the second on set invariance. We then illustrate how Lean can be used to analyse results from the literature, taking as example a classic lemma on the inversion of the noise filter in prediction-error identification (Ljung, 1999, Lemma 3.1). Since in Lean every step must be justified, the formalization brings to light a subtle point: the original claim of that lemma holds almost surely, although not for every realisation, as its statement may suggest. A simple counterexample makes this concrete. Finally, the paper discusses guidelines and perspectives on the use of Lean in system identification research.

Fri 13 NovArtificial Intelligence
The gist
Mathematical proofs in scientific research can be very long and complicated, making them hard to check for mistakes. The authors show how a software tool called Lean can write and check each step of these proofs automatically to ensure they are correct. They use examples from system identification, a control engineering field, to explain how this works. They also find that checking proofs carefully with Lean reveals subtle details that could be missed in traditional papers. Their work shows how using Lean can improve the reliability of math proofs in this area.
Open → 2611.18220v1

Verification method verifies continuous-time random processes reliably

A Hoare Logic for Continuous-Time Stochastic Processes

Abstract: Programming language researchers have developed a wide array of tools to reason about probabilistic programs, but it remains unclear how to apply these proof techniques to continuous-time stochastic processes. To bridge this gap, we propose a verification method for properties of continuous-time stochastic processes by reasoning about them as standard programs. To demonstrate our approach, we define a continuous semantics for a probabilistic imperative language $\mathrm{pWhile}^{\mathrm{dt}}$ by taking the Kuratowski limit of the standard semantics, which collects limit points of the standard program semantics as the time-step approaches zero. Then, we present a Hoare-style logic that uses loop invariants and decreasing functions to verify properties of continuous programs. To establish soundness, we identify a broad class of assertions that are preserved from the standard semantics to the continuous semantics. To demonstrate our approach, we verify properties of ten continuous-time stochastic processes modeled in $\mathrm{pWhile}^{\mathrm{dt}}$, including variants of Brownian motion and a stochastic hybrid system.

Mon 9 NovLogic in Computer ScienceProgramming Languages
The gist
Programs that include randomness over continuous time are tricky to analyze using usual methods designed for steps in time. The authors develop a new way to treat these continuous-time random processes like normal programs by considering what happens as time steps get really small. They create a set of rules, based on Hoare logic, to prove properties about such processes, including famous examples like Brownian motion. Their approach allows verifying that these processes behave correctly with respect to certain conditions.
Open → 2611.12761v1

Online learning stabilizes plasma controls with limited data

Online Imitation Learning for Stabilizing Vlasov--Poisson Plasmas

Abstract: We study the problem of controlling instability in the Vlasov--Poisson plasma systems, a fundamental challenge for nuclear fusion control. Motivated by the partial observability of plasma states in practice, we investigate imitation learning algorithms that learn from a fully-observable expert controller, but operate under macroscopic measurements. We theoretically establish a separation between online and offline imitation learning --- even a small behavior cloning error leads to error compounding that exponentially amplifies instability over time. By way of contrast, the online imitation learning loss can polynomially control the stability of rollout trajectories. We then propose a DAgger-style algorithm that learns the stabilizing controller under partial observability. Simulation results on a 1D Vlasov--Poisson system demonstrate that our algorithm can effectively stabilize plasma instabilities over longer time horizons than behavior cloning.

Sat 7 NovMachine Learning
The gist
Controlling unstable plasma, which is important for nuclear fusion, is hard especially when only limited information about the plasma is available. The authors studied how machines can learn to imitate a perfect controller that sees everything but must work using only limited data. They found that simply copying behavior leads to worsening performance over time, but interactive learning that keeps updating the controller can prevent this. Their new algorithm successfully keeps the plasma stable longer in computer simulations.
Open → 2611.10870v1

Quantum complexity of testing linear system stability analyzed

Quantum query complexity of Hurwitz stability testing

Abstract: Determining whether every trajectory of the linear system $\dot{x}=Ax$ converges to zero (or, equivalently, whether $A$ is Hurwitz stable) is a fundamental problem in control theory. It arises, for example, in verifying the stability of feedback controllers and state observers, as well as of linearized dynamical models. We study the quantum query complexity of deciding whether a matrix $A\in\mathbb{C}^{n\times n}$ is Hurwitz stable, given an exact normalized block-encoding of $A$. The query complexity is characterized in terms of the system dimension $n$ and a supplied lower bound $0<ν\leq ω_A$ on the imaginary-axis distance $ω_A:=\min_{t\in\mathbb{R}}σ_{\min}(A-\mathrm{i}tI)$, where $σ_{\min}(A)$ is the smallest singular value of $A$. We prove a query-complexity lower bound of $Ω(\sqrt{n}/ν)$, which holds even when $A$ is restricted to be real and diagonal. For general matrices with nonzero $ω_A$, we encode an approximation of the right-half-plane spectral projector $\operatorname{sign}^+(A):=(I+\operatorname{sign}(A))/2$ as a block of a matrix inverse, where $A$ is Hurwitz if and only if the projector is zero. To test this, we produce a quantum algorithm which determines if a $(β,b,1/(4\sqrt{n}))$-block-encoded projector matrix (like $\operatorname{sign}^+(A)$) is zero or not; for constant success probability, it has complexity $Θ(β\sqrt{n})$. When the zero test is applied to specific approximations of $\operatorname{sign}^+(A)$, this yields an upper bound of $O(\sqrt{n}ν^{-2}\log(\sqrt{n}/ν))$ queries in the general case, and $O(\sqrt{n}ν^{-1}\log n)$ when $A$ is Hermitian or normal. The latter upper bound matches the lower bound up to a logarithmic factor. Thus the dimension dependence of quantum Hurwitz testing is nearly optimal, while the remaining factor in the non-normal case isolates an open conditioning gap.

Fri 30 OctComputational Complexity
The gist
Checking if a system always settles down to zero over time—a property called Hurwitz stability—is important for control and engineering. The authors studied how efficiently a quantum computer can decide this stability for complex systems. They found mathematical limits on how many times the quantum algorithm needs to look at the data, depending on the system size and some technical details. They also designed an algorithm that nearly matches this limit, showing how quantum methods can quickly test system stability in certain cases. Some open questions remain for more complicated systems.
Open → 2610.41180v1

UPPAAL gains interactive help from large language models for system modeling

UPPAAi: Towards LLM-Assisted Modeling, Verification, Analysis and Synthesis

Abstract: For more than three decades, UPPAAL has supported the modelling and analysis of real-time and cyber-physical systems. Yet building a suitable model, formalising the intended properties, choosing analysis techniques and interpreting counterexamples remain iterative tasks that demand substantial expertise. Large language models (LLMs) can offer plausible explanations and even solutions, but they are not necessarily correct. We present UPPAAI, an extension of UPPAAL in which users construct, explain, debug and specify models in dialogue with an LLM. The guiding principle is that the user requests, the LLM proposes, explains and orchestrates, and UPPAAL verifies. UPPAAI exposes the UPPAAL editor, queries and verifier through a Tools API. We report on seven case studies, from explaining to a newcomer to learning a controller, covering timed automata, stochastic hybrid automata and strategy learning. LLMs proved to be extremely helpful in not just generating solutions but also pro-actively suggesting relevant UPPAAL features it was not asked about.

Thu 29 OctSoftware EngineeringArtificial IntelligenceLogic in Computer Science
The gist
Modeling complex systems like timed or hybrid automata can be hard and often requires expert knowledge. The authors created UPPAAI, a tool that integrates a large language model (LLM) with the existing UPPAAL software to help users build, explain, and check these models through a conversation. The LLM suggests and explains steps while UPPAAL verifies correctness, making model building more approachable and interactive. This new tool was tested on different case studies, showing that LLMs can proactively offer helpful features even when users don’t ask for them.
Open → 2610.40042v1

Python tool finds equations behind changing systems with confidence

py-ARGOS: A Python package for automatic, statistically principled discovery of ordinary differential equations from data

Abstract: Recovering the governing equations of a dynamical system from data is a central problem in data-driven science, yet the available Python tooling forces a compromise among automation, statistical rigor, and reliability. We present py-ARGOS, a Python package implementing two automatic algorithms, bootstrap ARGOS and Bayesian-ARGOS, behind a uniform interface. Both are statistically principled throughout: automated model selection reduces the candidate library, each coefficient is reported with a bootstrap confidence or Bayesian credible interval that quantifies its uncertainty, and a term is retained when this interval excludes zero. Built-in diagnostics additionally flag unreliable fits. On chaotic benchmark systems, the two algorithms improve on the SINDy family's identification performance.

Sun 25 OctMathematical Software
The gist
Understanding how things change over time often relies on finding equations that describe their behavior. The authors present py-ARGOS, a Python package that can automatically discover these equations from data, doing so in a way that carefully checks how certain the results are. It uses two smart algorithms that not only pick which parts of an equation matter but also show how sure we can be about each part. This helps avoid mistakes and works well even for chaotic or complex systems.
Open → 2610.34458v1

Polynomial system controllability cannot be fully decided or listed

Controllability of polynomial systems is undecidable and not recursively enumerable

Abstract: Sontag conjectured in 1988 that global controllability of polynomial control systems with rational coefficients is undecidable, but that the controllable systems form a recursively enumerable set. We prove the first part of the conjecture and disprove the second. For every homogeneous polynomial vector field $F$ of odd degree with rational coefficients, we construct an explicit polynomial control system $Σ_F$, with one more state than $F$, that is globally controllable if and only if the origin is \emph{not} globally asymptotically stable for $\dot x=F(x)$. By a recent result, the homogeneous polynomial vector fields with a globally asymptotically stable origin form a set that is recursively enumerable but not decidable. Consequently, global controllability of polynomial systems is undecidable, and the controllable polynomial systems are not recursively enumerable. The same conclusions hold for systems of the form $\dotξ=f(ξ)+Bu$ with a polynomial drift and a constant input matrix.

Mon 19 OctComputational Complexity
The gist
Determining whether certain mathematical systems called polynomial systems can be controlled is a complex problem that the authors study. They prove it is impossible to create a step-by-step method that always decides if these systems are controllable. Moreover, they show that you cannot even list all controllable systems with a systematic procedure. This settles a long-standing question about these systems' behavior.
Open → 2610.26832v1

Stability certificates for chains of delayed neural modules hold for any length

Length-Independent Stability and $L_2$-Gain Certificates for Chains of Delayed Neural Modules

Abstract: Networks of repeated neural modules with delayed neighbour coupling need stability guarantees for every number of modules, but delay-dependent criteria certify one network at a time and grow with its size. This paper derives stability and $L_2$-gain certificates of length-independent size for open chains and rings of modules of two alternating types with slope-restricted activations. A local and a transmission delay act independently and are only bounded and continuous. A Lyapunov-Krasovskii functional with fixed integration limits and module-local variables that repeat with the module type reduces the certificate of every chain to one cell of two modules. Local conditions and a single linear matrix inequality of order 96 on this cell, or 104 for the gain, certify global asymptotic stability and a common $L_2$-gain bound for all open chains and all rings with an even number of at least four modules. Up to a margin, the stability inequality is also necessary for the repeated variables to certify every length. All uniform certificates, including certified coupling regions, are verified in exact rational arithmetic. In the numerical study, the certified coupling reaches 92% of an outer bound given by unstable members of the model class for short delays and 79% at the nominal delay bounds. For 16 modules at unit coupling strength, certificates of the same functional computed for that length alone raise the certified delay scale by 2% to 7% in four cases. A single stability problem with 1298 scalar variables thus replaces per-length certification of long chains at little cost.

Sun 18 OctNeural and Evolutionary Computing
The gist
When neural networks are built by linking many identical modules with delays in their connections, it can be hard to prove they will always behave stably no matter how many modules there are. The authors developed a way to test stability using a fixed-size mathematical check that applies to any length of such chains or rings of modules. Their method also provides guarantees about how signals amplify through the network, making it easier to certify large systems without growing complexity. This approach reduces the problem of certifying very long networks to checking a small, repeating unit, verified with exact math.
Open → 2610.25147v1

Toric words improve checking system behaviors in noisy conditions

Model Checking for Noisy Toric Words

Abstract: Toric words have recently gained attention as a means to model the behaviour of dynamical systems such as linear loops. This work is concerned with describing when the toric model is precise enough to decide whether the trace of a system is contained in an omega-regular language. Our contributions are threefold. (i) We use topological means to generalise a class of toric words by defining noise-robust dynamical systems and their sets of noisy traces. We prove that all noisy traces of a system are almost-periodic, and have the same set of recurrent factors. (ii) We apply the abstract techniques above to concrete sequences obtained as sign descriptions of real algebraic linear recurrence sequences (LRS), and show that the language-membership problem for sign descriptions of LRS with few dominant roots is decidable. (iii) We show that noisy traces of noise-robust dynamical systems are indistinguishable to a prefix-independent omega-regular language, and inclusion is decidable provided the common language of their recurrent factors is recursive. We incidentally obtain properties of prefix-independent omega-regular languages that may be of independent interest.

Fri 16 OctFormal Languages and Automata TheoryLogic in Computer Science
The gist
Checking if complex systems behave as expected can be hard, especially when there is uncertainty or noise. This paper studies a special way of describing system behaviors called toric words and extends it to handle noisy data. The authors show that their noisy model keeps key repeating patterns that help in deciding whether a system’s behavior matches certain rules. They also prove it’s possible to decide membership in certain pattern languages for systems following these noisy models.
Open → 2610.23463v1

Seven distinct cycles found in a special forced equation model

Seven Limit Cycles in a Periodically Forced Three-Zone Equation

Abstract: We construct a seven-cycle example for a periodically forced scalar equation with a continuous three-zone piecewise-linear nonlinearity. Symbolic integration and linear elimination reduce the four-flight matching conditions to two equations in the residence times. The competing exterior visits produce an additional turn of the critical response and lead to the rational parameters $(A,B,μ)=(9,3/2,181363/100000)$. At these parameters, finite Taylor estimates give three sign changes of the return displacement. Scalar order, analyticity along transverse itineraries, and half-period reflection then establish at least seven distinct isolated periodic orbits on the phase cylinder.

Wed 14 OctSymbolic Computation
The gist
The paper shows that a particular kind of mathematical equation, which behaves differently depending on which of three zones it is in, can have at least seven repeatable cycles when it is forced periodically. The researchers used symbolic math to simplify the problem and found certain parameter values that produce multiple sign changes in the solution's return distance. These changes prove that at least seven separate periodic behaviors exist in this system. This helps understand the complex patterns possible in certain piecewise linear models.
Open → 2610.19586v1

Post selection inference improves reliability of sparse nonlinear dynamics discovery

PSI-SINDy: Post-Selection Inference for Sparse Identification of Nonlinear Dynamics

Abstract: Sparse identification of nonlinear dynamics (SINDy) is a data-driven framework for discovering governing dynamics from time-series data by identifying a sparse subset of candidate dynamical terms from a prespecified library. In this work, we develop a statistical inference framework for quantifying the reliability of dynamical terms selected by SINDy through hypothesis tests and confidence intervals. A key difficulty is that using the same noisy trajectory for both selecting dynamical terms and assessing their statistical significance can introduce selection bias. Post-selection inference provides a principled framework for addressing such bias, and we propose PSI-SINDy, a post-selection inference method tailored to SINDy. Direct application of existing post-selection inference techniques is challenging because SINDy involves measurement error in the candidate terms and shared noise between the response and design. To address these challenges, PSI-SINDy uses data thinning to decompose a single observed trajectory into four mutually independent views with distinct roles in selection and inference. This construction enables inference for selected dynamical terms while accounting not only for selection bias but also for measurement-error and shared noise effects. We establish the theoretical validity of PSI-SINDy under stated conditions and evaluate its performance through numerical experiments on simulated and experimental dynamical-system data.

Thu 8 OctMachine Learning
The gist
Discovering the underlying rules of how things change over time from noisy data is tricky, especially when you only want a few important rules. The authors worked on making sure that the rules chosen this way are truly reliable, not just lucky guesses from noisy measurements. They developed a method called PSI-SINDy that carefully checks these selected rules by cleverly splitting the data to remove bias and noise problems. This method was tested on simulated and real data and shown to be valid under certain conditions.
Open → 2610.11486v1

Latent lagrangian neural networks capture complex forced system dynamics

Latent-Lagrangian Neural Networks for Reduced Order Modeling of Non-autonomous Nonlinear Dynamical Systems

Abstract: This work proposes a latent Lagrangian-based framework for reduced-order modelling of forced nonlinear dynamical systems. In contrast with conventional Lagrangian or Hamiltonian neural networks, our approach learns a set of latent coordinates sufficient to capture the dynamics conjointly with two neural networks for the latent kinetic and latent potential energies, and leverages force supervision to eliminate the need for an ODE solver during training. Consistency of physical laws in the latent space is ensured through the principle of virtual work. Results show that the model effectively learns the subtle dynamics induced by the system's nonlinearity and non-convex potential energy, and generalizes well to unseen forces and initial conditions. These observations confirm the physical relevance of the proposed approach, and its interest for model reduction.

Sat 3 OctComputational Engineering, Finance, and ScienceMachine Learning
The gist
Modeling systems that change over time with outside influences can be very complicated. The authors developed a new way to simplify these complex systems into fewer variables while still keeping the important parts of the system's physics. Their method uses neural networks that learn energy patterns inside a hidden space and respects physical laws to predict system behavior without costly calculations during training. The resulting model can handle tricky, nonlinear behaviors and works well even when faced with new starting points and forces.
Open → 2610.04723v1

Ppo struggles to match analytical trading strategy under uncertainty

When a Correct Reward Is Not Enough: Diagnosing and Guiding PPO in an Analytically Solved Broker-Trader Game

Abstract: Reinforcement learning (RL) is increasingly used for financial optimal-control problems when complex dynamics make analytical strategies difficult to obtain. There are financial mathematics literactures which provides many solved models whose equations and controls could evaluate and guide learning; we ask whether RL can exploit these results. We place a proximal policy optimisation (PPO) agent in an analytically solved continuous-time broker--trader game. PPO replaces the broker and chooses its trading speed while interacting with an informed trader and stochastic uninformed order flow. We derive a finite-step reward from the broker's continuous-time payoff and verify its discrete implementation through grid refinement and an exact one-step identity. With zero uninformed flow, a validation-selected PPO--FFNN approaches the reference action. With stochastic uninformed flow, the tested PPO--FFNN and PPO--LSTM remain inaccurate, although supervised learning confirms that their actors can represent the action. Monte Carlo diagnostics show that their critics do not reliably rank nearby actions; potential-based reward shaping also gives no reliable improvement. Under partial information, a causal certainty-equivalent controller based on the broker's observable history remains close to the reference, while PPO has larger errors and lower payoffs. Finally, we freeze the analytical policy and train PPO to adjust it after the execution cost changes. Halving the cost yields a repeatable improvement that closes \(2.22\%\) of the gap to the changed-cost reference. The analytical solution therefore provides both a benchmark for diagnosing RL and a useful starting policy for adaptation.

Fri 2 OctArtificial Intelligence
The gist
The paper looks at how a common reinforcement learning method, called PPO, performs in a financial trading game where the best strategy is already known through math. When there is no random uncertainty, PPO can learn to act like the optimal strategy. But when there is randomness in the market, PPO doesn't perform well because its way of evaluating possible actions isn't reliable. The authors also find that starting PPO from the known solution helps when the market changes slightly.
Open → 2610.03598v1

New neural network models systems with multiple stable states

Port-Hamiltonian Neural Networks for Systems with Multiple Asymptotically Stable Equilibria

Abstract: Stable port-Hamiltonian neural networks certify asymptotic stability by construction. Yet, their Hamiltonian is a global Lyapunov function with a single global minimum, so they can represent only dynamic systems with {one} attractor. We demonstrate that this excludes even simple systems with energy landscapes forming a double well, and we overcome the restriction by parametrising the Hamiltonian as a {product} of Bregman divergences generated by one input-convex network. We prove that the resulting model is locally Lyapunov stable, that the coexistence of stable equilibria forces additional non-asymptotically-stable equilibria to exist, that all equilibria lie in a bounded region, and under a hyperbolicity assumption that almost-everywhere stability holds. On three systems our approach is able to recover the energy surface characteristics and improve the convergence speed by 1.8$\times$-8.5$\times$.

Thu 1 OctMachine Learning
The gist
Some physical or engineered systems can settle into different stable conditions, but existing neural network models only handle one. The authors find that current methods miss system behaviors like two stable points and fix it by designing neural networks that represent multiple valleys of energy. They prove their approach is mathematically sound, showing all stable states stay within a certain range and that the system behaves well in most cases. Their method works better than previous ones on example problems, making predictions converge faster.
Open → 2610.01356v1

Linear differential equations with rational coefficients can be found efficiently

Rational linear forms of linearizable ordinary differential equations

Abstract: Every scalar ordinary differential equation of order at least three with rational right-hand side that is locally linearizable by a point transformation admits a linear form with rational coefficients over the same coefficient field. We prove this by restricting the derived symmetry algebra to a coordinate line and recovering a scalar differential operator from rational symmetry-jet data. The construction uses differential elimination and linear algebra; it does not require the symmetry generators or a linearizing transformation to be solved for. A single integer parameter suffices to choose the line.

Thu 1 OctSymbolic Computation
The gist
The paper shows that certain complicated equations involving derivatives can always be rewritten in a simpler, linear form if their terms come from rational functions. The authors provide a way to find this simpler form without directly solving the usually hard problem of finding symmetry or transformations. Their method uses algebra and calculus tools to do this with just one parameter choice, making the process more manageable.
Open → 2610.01338v1

Quantum methods enable projections onto nonorthogonal eigenspaces

Quantum oblique eigenprojection

Abstract: Every square matrix decomposes its underlying Hilbert space into generalized, nonorthogonal eigensubspaces. We show that a quantum computer can perform such an oblique eigenprojection $Π$ given block encoding access to the input matrix. Our approach has a query complexity nearly linear in the inverse gap and a normalization factor close to $\lVertΠ\rVert$ under a spectral-set condition on the input. This covers common assumptions on the numerical range or diagonalizability and matches known results for orthogonal eigenprojections. We achieve this with a two-sided block preconditioning that uses a discrete Fourier transform of the matrix resolvent. We describe applications to: (i) preparing eigenstates of matrices with complex eigenvalues, extending the quantum eigenvalue transformation algorithm of Low and Su beyond real spectra; (ii) solving continuous-time algebraic Riccati equations, cubically speeding up a prior solver of Rodenas-Ruiz, Zhao, and Lee; and (iii) solving ordinary Sylvester equations, quadratically improving a direct augmented method of Wang and Liu. Our result suggests a promising route to applying nonanalytic matrix functions on quantum computers.

Wed 30 SeptData Structures and Algorithms
The gist
Complex systems can be understood by breaking them down into parts described by matrices, but these parts don't always fit neatly and orthogonally. The authors show how a quantum computer can create projections onto these angled, or oblique, parts of a matrix's space. This extends earlier quantum techniques restricted to simpler, orthogonal parts and can speed up solving certain matrix equations. These advancements may help apply more complex matrix functions on quantum machines.
Open → 2609.40254v1

Hyperreservoir networks improve prediction of changing time series

Context-dependent time-series prediction via HyperReservoirs

Abstract: Time series prediction is a common application of reservoir computing. When the training and testing time series data contains multiple dynamical regimes, because an underlying parameter is changing, or the data in fact consists of multiple distinct systems, simple application of the reservoir computing principle produces high prediction errors. Here, we propose a HyperReservoir as an extended model of reservoir computing especially designed for such cases. The HyperReservoir combines a main reservoir with a smaller context reservoir, where the latter modulates the output weights of the former. This structure resembles the hypernetworks from deep neural network literature. However, in contrast, HyperReservoirs retain the simple training via linear regression of standard reservoir computing. We compare the proposed architecture with a conventional ESN, in which context acts at the input, and a full-matrix Conceptor, in which context modulates the reservoir state space. We evaluate all three models on time-series prediction tasks based on Lorenz and Rössler systems, including for varying bifurcation parameters and time sampling scales. We find that the HyperReservoir achieves the lowest mean test error in all three tasks, and particularly outperforms conceptors on data that is sampled from the same attractor but at different time scales.

Mon 28 SeptMachine Learning
The gist
Predicting future data points in changing or complex time series is hard because the underlying rules can shift. The authors introduce HyperReservoirs, a special neural network design that uses one part to learn the data and another smaller part to adjust the prediction based on context. This approach keeps training simple but adapts better than older methods. Tests on well-known chaotic systems showed HyperReservoirs predicted future points more accurately than competing models, especially when the data changed speed but not overall behavior.
Open → 2609.34847v1

Neural network observer trains faster with provable stability guarantees

Learning Provable Neural Network Observer for Uncertain Dynamical Systems

Abstract: In many safety-critical applications, control of uncertain dynamical systems relies on observers that estimate states and external disturbances. Neural network observers can improve estimation accuracy, but certifying their Lyapunov stability via Linear Matrix Inequality (LMI) constraints leads to large-scale semidefinite programs (SDPs) that are difficult to solve for large networks. To overcome this scalability bottleneck, we propose a novel two-stage training framework for provably stable neural network observers. Our approach decouples the optimization into a point-guided Lyapunov pre-training phase, which rapidly achieves high estimation accuracy and local stability over sampled states, followed by an LMI fine-tuning phase that efficiently satisfies a strict global Lyapunov stability certificate. We provide formal theoretical guarantees for local stability radii and probabilistic coverage over a prescribed compact error-state domain under specified regularity and sampling assumptions. Experiments on nonlinear control benchmarks and X-29 aircraft ablations show that our LMI-certified neural network observers train significantly faster than direct LMI-based methods and generalize robustly across diverse systems, achieving improved tracking accuracy over a range of observer baselines. The code is available at https://github.com/Berry-Myon/LearningNeuralNetworkObserver.

Fri 25 SeptMachine Learning
The gist
Controlling systems that change over time and face uncertainties requires ways to estimate current states accurately. The authors developed a new two-step training method for neural networks that observe and estimate these states more reliably and faster than before. Their approach first quickly learns to estimate well in specific cases, then fine-tunes to ensure the system remains stable overall. They prove that their method guarantees stability within a certain range and demonstrate better accuracy on various control tasks including aircraft models.
Open → 2609.30819v1

Quadratic limits found for uncompletable words and zero matrix products

Quadratic bounds for uncompletable words and matrix mortality

Abstract: Every finite nonempty incomplete uniquely decipherable code with maximum word length $k$ has an uncompletable word of length at most $4k^2-3k$. The bound is independent of the number of codewords and their total length. Deleting a complete codeword cycle gives a finite path-counting identity; Kraft equality then supplies a short word of deficient compressed mass. Cyclic averaging and padding turn it into an uncompletable word. Conditional expectation makes the construction polynomial-time and also decides completeness. First-return words extend the bound to mortal families of nonnegative integer $n\times n$ matrices with joint spectral radius at most one, provided every strongly connected component has a vertex meeting every cycle. Such a family has a zero product of length at most $4n^2-3n$. A binary partial deterministic family with $2k-1$ states has shortest zero product of length $k^2+k-1$, establishing the optimal quadratic order. The bounds and the explicit-code algorithm, including its polynomial work bound, are proved in Lean.

Fri 25 SeptFormal Languages and Automata Theory
The gist
Some sets of code words can’t be completed to represent all possible messages, and finding a short example of such uncompletable words helps understand this limitation. The authors found a mathematical rule that the shortest uncompletable word in such cases is at most about four times the square of the maximum word length, which doesn’t depend on how many code words there are. They also extended this idea to certain collections of matrices, showing that you can multiply some of these matrices a limited number of times to get a zero result, with rigorous size limits. They made their proofs and algorithms fully precise and checked in a formal verification system called Lean.
Open → 2609.30817v1

Next generation reservoir computing infers missing parts of complex systems

Inference of Unknown Dynamical Components Using Next Generation Reservoir Computing: From Chaotic Systems to Climate Data

Abstract: We investigate next generation reservoir computing (NGRC) as a data-driven approach for inferring unseen components of dynamical systems. We compare NGRC with traditional reservoir computing (RC) using the Lorenz and Rössler system, where two unknown components are inferred from one given component. For both systems, NGRC achieves accurate results while requiring fewer training data and less computational time than RC. We identified an inverse proportional behavior between the number of time-delayed steps needed for NGRC and the temporal resolution, indicating that the physical time span covered by the delay interval is an important factor in determining the required number of delayed steps. Finally, we apply NGRC to the observational climate data of ENSO (El Niño--Southern Oscillation) and infer one observable from the remaining variables. Despite the noise and complexity of the real-world data, the NGRC shows promising results. Our findings demonstrate the potential of NGRC for efficient inference of unseen components in both controlled dynamical systems and real-world data.

Mon 21 SeptMachine Learning
The gist
Many systems, like the weather or climate, have parts we cannot directly observe, making prediction challenging. The authors used a method called next generation reservoir computing to guess missing parts of these systems from available data. They tested this on well-known chaotic systems and real climate data, showing it works well using less data and time than older methods. This approach might help better understand and predict complex natural phenomena by filling in unknown details.
Open → 2609.24754v1

Optimizing stability certificates improves safety areas for polynomial systems

Optimizing Lyapunov Certificates via Stability-Preserving Quadratization for Polynomial Systems

Abstract: Region-of-attraction (ROA) certificates for polynomial systems become expensive as state dimension and degree grow: direct sum-of-squares (SOS) formulations require combinatorially growing monomial bases. Quadratization represents a polynomial vector field exactly on an invariant manifold of a quadratic system, allowing a quadratic Lyapunov function to certify the ROA. For a fixed lift, stabilizer gains shape the off-manifold extension and transverse dynamics, while representation gauges change the matrix representation but not the vector field. Both affect the spectral-norm certificate, yet prior work fixes the gain by a feasibility heuristic before optimizing the gauge. We formulate optimal dissipative quadratization (ODQ), jointly designing gains and gauges for a fixed monomial lift, reference extension, stabilizer factorization, and Lyapunov weight $Q=I$. Gains lie in a prescribed compact Hurwitz box. At each gain, an exact semidefinite program globally minimizes the spectral-norm bound over the gauge. Residual-aware bounds yield a certified closed Lyapunov sublevel set, accounting for the floating-point Lyapunov residual. Under our stated assumptions, every accumulation point of the idealized outer search is box-Clarke stationary. A finite run returns the best independently verified candidate; global optimality of the gain search is not claimed. On a planar quintic, optimizing the gain increases the certified area by a factor of $2.238$ over a matched zero-gain gauge. Across 16 heterogeneous polynomial systems with stabilizer freedom, ODQ improves on both fixed-gain lifted baselines. All 36 ODQ runs on the relay benchmark complete, and all 27 repeat-level comparisons across nine fully paired cases favor ODQ over an SOS baseline with a fixed quadratic Lyapunov function in both the fixed-direction proxy and construction time. Broader comparisons with direct SOS methods remain mixed.

Wed 16 SeptSymbolic Computation
The gist
Checking how stable a system is—such as predicting if it will stay safe over time—can get very complex for systems described by polynomial equations. The authors studied a way, called quadratization, that simplifies this checking by turning complicated equations into quadratic ones, which are easier to handle. They improved the method by jointly tuning certain parameters, called gains and gauges, that affect the accuracy of verifying stability. Their approach finds better settings that certify larger safe regions compared to previous methods, especially shown in tests on multiple example systems.
Open → 2609.18871v1

Large language models steady networked control systems with slow supervision

Large Language Models in the Loop: A Stability- and Network-Aware Survey in Networked Control, Cyber-Physical, and Multi-Agent Systems

Abstract: Modern networked control systems (NCSs), cyber-physical systems (CPSs), and complex multi-agent network systems (CNSs) increasingly rely on large language models (LLMs) for high-level decision-making. However, the slow, stochastic nature of LLMs directly conflicts with the strict stability and safety guarantees required by these physical systems. This survey presents a unified analysis of how LLMs can be admitted into the control loop of NCS, CPS, and CNS without compromising closed-loop guarantees. We organize this around a core principle: the LLM operates as a slow supervisor adjusting high-level goals and constraints, while a fast, certified inner loop maintains physical stability. Under this framework, LLM integration maps directly to classical networked control challenges, where inference latency acts as delay, API failures as packet dropouts, tokenization as quantization, and hallucinations as bounded disturbances. We assess current developments across all these three domains, highlighting that rising model capabilities are frequently accompanied by a drop in formal safety assurances. Finally, we propose concrete future research directions, identifying the widespread lack of formal stability proofs as the field's central open problem.

Tue 15 SeptArtificial Intelligence
The gist
Some computer systems that control physical devices or multiple agents need to be very reliable and stable. The paper's authors review how large language models (LLMs), which usually take time to respond and can be unpredictable, can still be included in these systems safely. They explain that if the LLM acts slowly to set high-level goals while a faster, certified system handles immediate control, stability can be maintained. They also compare typical LLM behaviors to known network control problems and say more research is needed to guarantee safety formally.
Open → 2609.16599v1