Papers

We study a continuous-time $K$-unit online resource allocation problem with nonhomogeneous Poisson arrivals and time-varying valuation distributions. While the optimal dynamic policy generally depends on both the remaining inventory and the time left in the horizon, we focus on a simpler and practically appealing class of stock-based threshold policies, whose limited number of thresholds depend only on the number of units remaining. We evaluate these policies against the multi-unit prophet benchmark, which selects the best $K$ realized values in hindsight. Our main contribution is a new competitive-analysis framework for stock-based thresholds in continuous time. We first reformulate the problem through a type-covering dual. The central challenge for analyzing the dual is that the dual is both infinite-dimensional and non-convex: the adversary can choose time-varying arrival and valuation processes, while the policy performance depends nonlinearly on the stochastic inventory trajectory. We overcome these challenges by reducing the continuous-time adversarial problem to a Poisson optimization $PoisOPTRe_K$, and then proving sharp structural properties of its worst-case solutions. In particular, adversarial arrivals admit cutoff and late-filling structures, which yield an exact four-parameter formulation for two thresholds and a finite nested-interval representation for general thresholds. These reductions make the guarantees directly computable. For example, for two thresholds, we obtain a competitive ratio $0.6269$ for $K=2$; with three thresholds, we obtain the ratio $0.6816$ for $K=3$. In this way, we show that simple stock-based thresholds achieve strong prophet-inequality guarantees despite ignoring calendar time.

Large-scale optimal transport (OT) problems involve a vast number of transport variables, leading to prohibitive memory and computational costs. To address these challenges, we propose a multiscale primal-dual interior-point relaxation method (MSIPRM). The multiscale outer framework constructs a hierarchy of standard OT problems at progressively finer levels. At each level, the OT problem is solved over a sequence of adaptively refined active sets initialized based on the solution support at the previous level. This yields a sequence of closely related sparse subproblems, thereby substantially reducing memory requirements. The primal-dual interior-point relaxation method (IPRM) serves as the inner solver for each sparse subproblem. Since IPRM does not require strictly interior iterates, it can readily use the solution of the previous subproblem as a warm start. To efficiently obtain the Newton direction, we solve a reduced Schur complement system derived from the normal equations. Furthermore, we develop an effective support-identification strategy based on the approximate solutions obtained by IPRM. We establish condition number estimates for the Schur complement matrices and analyze the global and local convergence properties of the algorithm. Numerical experiments on large-scale test problems demonstrate the computational efficiency and scalability of MSIPRM and show that it compares favorably with existing solvers. In particular, MSIPRM can handle instances whose full formulations contain trillions of transport variables.

Reconstruction of 3D electromagnetic currents from sparse, multi-frequency, far-field radiation data is severely ill-posed because sparse sampling masks specific current Fourier modes and the missing spectrum enlarges the null space of the forward operator. Although structured-Hankel completion can recover missing spectrum by exploiting the finite rate of innovations structure of a compactly supported geometrically sparse source, it treats current components independently without guaranteeing compatibility with the model. This often renders non-physical solutions that violate model consistency. To resolve this, we propose a training-free Maxwell-model-consistent 3D structured-Hankel framework (MC-Hankel) that alternates joint low-rank Hankel completion with an exact, closed-form projection onto the true Maxwell synthesis subspace while enforcing conjugate symmetry and noise-aware data fidelity. Across tests on two source models, three sparse sampling rates (30%, 40%, and 50% of a Nyquist-sampled grid), and clean versus noisy (10 dB SNR additive white Gaussian noise) conditions, MC-Hankel consistently outperforms sub-sampled Fourier inversion, scale-normalized $\ell_1$-compressed sensing baseline, and the standard joint 3D annihilating filter-based low-rank Hankel matrix completion approach (ALOHA). Across all considered source-noise-sampling configurations, MC-Hankel improves mean peak SNR (PSNR) by 0.65-4.22 dB and tri-planer structural similarity (3D SSIM) by $0.0266$-$0.0703$, reduces relative full-volume $\ell_2$ reconstruction error by $7.4\%$-$38.9\%$, and brings non-physical model residuals down to machine precision over standard joint 3D ALOHA with merely 17% computational overhead. Exact signed-rank tests are underpowered for five paired trials, so statistical conclusions are reported together with paired bootstrap intervals, effect sizes, and small-sample limitations.

Acceleration for deterministic root-finding problems has been extensively studied in recent years; specifically, the anchor-based, or Halpern-type methods achieve optimal convergence rates with respect to the operator norm. However, acceleration via these methods does not directly carry over to stochastic setting due to accumulation of errors, unless one enforces diminishing variance via increasing batch sizes or variance reduction techniques. In this work, we show that another class of acceleration, namely the dual-anchor mechanism, extends to the stochastic setting without such error accumulation, in contrast to anchor-based algorithms. Consequently, we cleanly achieve $O(\epsilon^{-3})$ complexity with iteration-independent batch size, without any variance reduction or double-loop recursive regularization, for stochastic root-finding (resp. fixed-point) problems with cocoercivity (resp. square-nonexpansivity) in expectation. For strongly monotone operators, the same algorithm attains a sharper $\widetilde{O} (\epsilon^{-2})$ complexity, nearly matching the lower bound in terms of $\epsilon$-dependence.

Bregman proximal stochastic gradient (BPSG) methods bring variance-reduced composite optimization to objectives whose geometry is poorly captured by Euclidean smoothness. Their performance, however, remains sensitive to the step size: raw stochastic curvature estimates can fluctuate sharply, whereas line searches add repeated proximal evaluations. We introduce Ada-BPSG, a line-search-free BPSG method that couples the SAGA gradient table with a stabilized Barzilai–Borwein (BB) candidate. A mediant aggregates incremental secant information so that nearly singular local ratios receive little weight, and an explicit safeguard translates the resulting curvature estimate into the bounded step-size sequence required for convergence. This design yields a direct analytical chain from relative smoothness and component-wise variance control to convergence in finite-dimensional normed spaces. We prove an $O(n/K)$ ergodic rate for convex objectives, a restarted linear rate under relative quadratic growth, and an $O(1/K)$ bound for a Bregman proximal residual in the nonconvex setting. On logistic regression and sparse nonnegative matrix factorization, Ada-BPSG combines low objective values with substantially less sensitivity to the initial step size than standard variance-reduced baselines, while avoiding line search.

In this paper, we establish variational formulas and a rearrangement inequality for principal eigenvalues of asymmetric second-order difference operators on periodic lattices. For a general irreducible nearest-neighbor operator, positive right and left eigenvectors give an explicit saddle point and hence equal minimum–maximum and maximum–minimum formulas over positive profiles and periodic logarithmic correctors. The corrector is unique and satisfies a nonlinear discrete flux-conservation law. For exponentially tilted symmetric diffusion, the formula separates the discrete Dirichlet and potential terms from an exact nonlinear periodic correction. In one dimension the flux is constant, and an elementary scalar-flux representation yields a bell-shaped cyclic rearrangement that maximizes the tilted principal eigenvalue for every tilt.

Inverse Optimal Control (IOC) aims to infer the underlying cost functional of an agent from observations of its expert behavior. This paper studies the finite-horizon continuous-time inverse LQR problem from closed-loop state–input trajectories, where both the system matrices and the quadratic cost are unknown. The finite horizon induces a time-varying optimal gain, and this endogenous excitation serves as the structural mechanism that makes joint recovery possible. We quantify this mechanism through three computable conditioning indices, which measure state richness, gain-variation richness, and injectivity of a structured cost operator. Using these indices, we establish joint identifiability conditions for the inverse problem considered here. Crucially, these conditions guarantee recovery of the ground-truth system matrices $(A,B)$ and the true cost weighting matrices, rather than merely a behaviorally equivalent surrogate. We also develop a conditioning-aware sampled-data reconstruction method that reconstructs the gain $K(\cdot)$ and the closed-loop dynamics matrix $A_c(\cdot)$ from noisy measurements, recovers $(A,B)$ in closed form, and identifies the quadratic weights through a convex semidefinite program. We further establish the non-asymptotic perturbation bounds and the consistency of the full reconstruction method under sub-Gaussian observation noise, with explicit dependence on the same conditioning indices. Numerical experiments support the theory and illustrate the diagnostic value of the conditioning indices.

This paper presents a data-driven stable manifold (DD-SM) method, which integrates Koopman operator representation learning with the geometric stable manifold approach to Hamilton-Jacobi-Bellman (HJB) equations, enabling end-to-end optimal feedback control synthesis from raw trajectory data without prior knowledge of system dynamics. We construct an augmented control system under a unified symmetric subspace decomposition (SSD) and extended dynamic mode decomposition (EDMD) framework for joint approximation of the drift field, control matrix and their spatial derivatives, and derive probabilistic finite-sample error bounds for invariant and non-invariant dictionary spaces to yield a provably accurate approximate characteristic system of HJB equation. Via Lyapunov-Perron operator and ODE perturbation analysis, we prove the data-driven stable manifold achieves monotonically decreasing semi-global error with growing training data. We further establish closed-loop exponential stability and quantify the optimality gap, both tightenable by refining model accuracy. An efficient algorithm pipeline with adaptive data generation and deep neural approximation is developed, outputting control signals within 1 millisecond. Experiments on a modified van der Pol oscillator verify the effectiveness of our method.

We revisit the Sinkhorn-Knopp (SK) algorithm for the matrix scaling problem. Despite extensive literature on the global convergence of SK and its variants, its local linear convergence behavior remains less understood. We address this gap by providing the first nonasymptotic local analysis of SK that matches the rate obtained from existing asymptotic Jacobian-based arguments. We show that under certain connectivity conditions, SK is a polynomial-time algorithm for doubly stochastic matrix scaling. With the developed tools, we showcase the local suboptimality of SK and provide accelerated variants. Finally, for dense matrices, we improve the complexity of existing first-order matrix scaling algorithms from $O(\tfrac{n^{7/3}}{\varepsilon^{2/3}})$ to $O(\tfrac{n^{9/4}}{\sqrt{\varepsilon}})$.

This paper proposes a Wolfe-type spectral conjugate gradient method for nonsmooth convex optimization, built on the Moreau-Yosida regularization of the objective function. The method combines a safeguarded spectral parameter with a Dai-Kou-type conjugate parameter, and uses a Wolfe-type line search compatible with the inexact gradients that the regularization produces. We establish global convergence of the method, together with an R-linear convergence rate under an additional strong-convexity assumption. The method is evaluated on standard nonsmooth optimization benchmarks and on large-scale problems, and compared against several existing conjugate gradient and bundle-type methods. The results show that the proposed method performs competitively overall, matching or outperforming existing methods on most problems tested, while a few specific limitations of the current implementation are also identified and discussed

A metric space $Z$ gives rise to three natural classes of infinite-dimensional metric spaces associated to $Z$: $p$-Wasserstein spaces of probability measures on $Z$, nonlinear Lebesgue $L^p$-spaces of $Z$-valued maps, and $p$-Gromov-Wasserstein spaces of $Z$-valued kernels. The latter class, referred to as $Z$-Gromov-Wasserstein ($Z$-GW) spaces, extends the classical Gromov-Wasserstein framework from metric measure spaces to more general, possibly attributed, network-like structures, and unifies many GW-type distances that nowadays play a significant role in metric geometry, data science and machine learning. In this article we develop a unified metric-geometric theory of these three classes of spaces, with a particular focus on the $Z$-GW spaces. Our first main result identifies a fundamental submetry structure linking them: the nonlinear Lebesgue space maps via a submetry onto the $Z$-GW space, which in turn maps via a submetry onto the Wasserstein space. This structure provides a mechanism for transferring geometric information among the three spaces. We apply this framework to geodesics and Alexandrov curvature. For $1<p<\infty$, we prove that geodesicity of $Z$ is equivalent to geodesicity of each of the three associated spaces; in the endpoint case $p=1$, all three associated spaces are geodesic, even when $Z$ is not. We also characterize geodesics in the $Z$-GW space as generalized interpolations, extending a known characterization in the classical setting due to Sturm. Finally, we give a complete classification of Alexandrov curvature bounds for these spaces in terms of the curvature of $Z$. Thus, while the main focus of the paper is a new metric-geometric theory of $Z$-GW spaces, the submetry framework also extends classical theorems for Wasserstein and Gromov-Wasserstein spaces and yields new geometric consequences for nonlinear Lebesgue spaces.

One of the basic problems in mathematical learning theory is to identify a function $f: \Omega\to {\mathbb R}$ with certain specific properties which fits a given training data $\{(x_i, \xi_i)\in \Omega\times {\mathbb R}: i=1, \ldots, n\}$, in the sense that, $f(x_i) = \xi_i$ for $i=1, \ldots, n$, where $\Omega$ is a compact subset of ${\mathbb R}^d$ for some $d\in {\mathbb N}$ and $f$ is required to have certain specified characteristics. We address this problem when $f$ belongs to an arbitrary normed linear space ${X}$, and the evaluation maps $f\mapsto f(x_i)$ are replaced by maps of the form $ f\mapsto \varphi_i(f)$ on $X$, where $\varphi_1, \ldots, \varphi_n$ are continuous linear functionals on ${X}$. Using an inner product structure on ${\mathbb R}^n$, we shall device a method of least-squares for obtaining an approximate solution for the above problem and also identify a subspace of ${X}$ in which the least-square solution is unique, which in turn is shown to be equivalent to solving a matrix equation. In the context when the matrix under consideration is ill-conditioned, a regularized equation is devised, and order optimal error estimates are derived for the exact targeted data $\xi=(\xi_1, \ldots, \xi_n)$ and also when it is noisy, by choosing the regularization parameter appropriately.

We study the implementable finite-batch particle algorithm for mean-field variational inference as a fully discrete stochastic approximation of the projected Wasserstein dynamics. The target potential is globally smooth but need not be strongly convex. The departure from contractivity is quantified by the curvature defect \[ \mathfrak d_\alpha(x,y) = \bigl[\alpha\|x-y\|^2- \langle\nabla V(x)-\nabla V(y),x-y\rangle\bigr]_+, \] which is the additive loss in the one-step Euler contraction estimate. We prove a non-asymptotic Wasserstein stability bound that separates initialization, product-empirical approximation, finite-batch drift error, time discretization, and the defects accumulated along the coupled trajectories. Under the uniform bound $\mathfrak d_\alpha\leq\beta$, the particle iterates remain within $O(\sqrt{\beta/\alpha})$ of any MFVI minimizer, up to explicit errors in the particle number, batch size, and step size. The proof uses a stationary comparison array whose population law is an MFVI minimizer but whose particle-level law is a random product empirical measure, and it controls the resulting projected-drift discrepancy explicitly. We also give coordinatewise defect estimates and structural conditions for dimension-independent projected-drift sensitivity, construct an arbitrary-dimensional smooth nonconvex benchmark with a closed-form MFVI minimizer, and explain why polynomially growing drifts require a modification of the untamed explicit scheme.

We provide an estimator for the perturbed utility route choice (PURC) model that works with data at the level of individual trips. The estimator is a nested fixed-point algorithm that combines an upper bias-corrected linear regression problem with a lower individual-level perturbed utility maximization problem. We establish the statistical properties of the microPURC estimator and confirm these results with an experiment using simulated data. Finally, we demonstrate the estimator in practice using a large real-world dataset.

We study the capacity of a memory-based quantum repeater in entanglement swapping between two quantum links with either single or multiple memories, which we refer to as the end-to-end (E2E) entanglement throughput, subject to a constraint on the minimum fidelity. In order to approximate the E2E entanglement throughput, we adopt queueing models, where quantum links can have different characteristics: memory capacities, entanglement attempt rates and success probabilities, as well as classical communication latencies. We develop a model for estimating E2E entanglement fidelity, while taking into account the heterogeneous dephasing and depolarizing dynamics of quantum memories and Bell-state measurements in entanglement swapping as well as classical communication delays and noises. Finally, with the help of our models for approximating the E2E entanglement throughput and fidelity, we use the maximum waiting times of entanglements in quantum memories at the repeater as optimization variables to maximize the E2E entanglement throughput while ensuring required minimum E2E fidelity.

The Performance estimation problem (PEP) approach reformulates finding the exact worst-case performance of an algorithm as the solution to an optimization problem. Tractable formulation of the problem requires necessary and sufficient interpolation conditions. We present the interpolation conditions for convex–concave functions, bilinear functions, and composite functions with bilinear coupling. We also construct PEP for first-order methods for composite saddle-point problem.

Public Transport (PT) lines are traditionally designed to optimize performance under nominal traffic conditions. In practice, operating conditions frequently deviate from nominal ones, leading to substantial performance deterioration. Existing adaptation mechanisms typically rely on reactive interventions, such as stop-skipping, which are insufficient under large or recurrent traffic fluctuations. In such contexts, incremental adjustments may not suffice. This paper evaluates the potential of deeper structural redesigns to preserve performance. We propose a method to proactively redesign appropriate parts of PT networks under high traffic fluctuations that would otherwise deteriorate operator and user performance. We adopt a predict-then-optimize paradigm in which PT lines are reconfigured based on traffic forecasts using the Non-dominated Sorting Genetic Algorithm III (NSGA-III). To ensure operational feasibility and avoid excessive structural changes, we enforce high Jaccard edge overlap between the original and redesigned networks. To assess prediction inaccuracies, we construct a statistical model of errors from a well-established deep learning predictor, the Diffusion Convolutional Recurrent Neural Network, trained on real-world data. Computational results on Mandl's benchmark and the large-scale Beijing network show that controlled PT line redesign yields substantial user-centric performance gains and operational cost reductions under high traffic fluctuations while limiting topological changes. On the Beijing network under high variability, average travel time improves by up to 25.8% while preserving over 85% line overlap. Unlike stop-skipping baselines, which break connectivity for many OD pairs, our redesign preserves full OD connectivity. These results support a shift from static planning toward continuous and adaptive PT network design.

We study new algorithms for Contextual Bandits with Knapsack. In these problems, there are finitely many types of customers, products, and resources. Each product is made from a fixed combination of resources, and resources have finite capacity. A decision maker must assign each arriving customer one out of a set of multiple possible products. Every assignment of a customer to a product will generate a random reward, which equals an unknown linear function of customer and product features, plus a noise term. The objective is to jointly learn the mean reward function, and to make online assignments to minimize the expected revenue loss relative to an optimal policy that knows the reward function. We propose a natural and simple extension of the Upper-Confidence-Bound (UCB) family of algorithms and apply re-optimization techniques. We show that by taking advantage of re-optimization, our algorithm achieves an average regret of $O(\frac{(\ln T)^3}{T})$ where $T$ is the horizon length. Our bound significantly reduces the $O(\frac{1}{\sqrt{T}})$ bound in the literature for closely related dynamic-pricing problems that are based on re-optimization.

The time-optimal solution for a double integrator is a foundational result: its time-displacement law $T_r = 2\sqrt{L}$ for a normalized step $L$ is structure that numerical methods cannot provide. We solve the time-optimal control for rest-to-rest maneuvers of a double integrator coupled to a harmonic oscillator, deriving the first analytical solution and closed-form time-displacement law. Synthesis reduces to a single scalar inversion. This is the model to which the two-mass-spring system, the one-bending-mode flexible structure, and the linearized overhead crane reduce. A Pythagorean identity $T^2 = T_r^2 + 2T_s^2$ decomposes the optimal maneuver time into the rigid-body minimum and a synchronization time $T_s \leq \pi$, the cost of flexibility, giving the sharp envelope $2\sqrt{L} \leq T \leq 2\sqrt{L + \pi^2/2}$. The penalty is scale-dependent. For small maneuvers $T \propto L^{1/4}$: one oscillator mode costs as much as two additional integrators. For large ones, flexibility is asymptotically free. It vanishes at the natural motions, where the oscillator completes whole cycles. A closed-form sensitivity formula proves the maneuver time non-monotone in stiffness. Adding flexibility to a rigid body never shortens a maneuver, yet softening an already flexible structure can: stiffer is not always faster. The natural motions are robust design targets since first-order sensitivity to stiffness vanishes there.

This paper proposes a contextual quality-diversity evolutionary reinforcement-learning controller, CQD-ERL, for the supervisory control of a tropical, water-cooled chiller plant and its associated air side. Rather than converging to a single scalarised policy, the controller maintains a product archive of specialised policies indexed jointly by a data- driven operating context, a cluster of daily weather and load regime, and a context-invariant behaviour descriptor, filled by a gradient-free evolutionary operator and a soft-actor-critic policy-gradient operator that share one replay buffer. Every action is filtered through a deterministic safety shield before execution. The controller is trained on a two-tier reduced-order environment representing the latent load, cooling-tower approach and humidity constraints of a Singapore commercial building, and is evaluated over a full annual backtest against an ASHRAE Guideline 36 baseline.

Basin is a numerical optimization library for the Rust programming language. Numerical optimization is the task of finding the inputs that minimize a function, and it is a fundamental element across the sciences: fitting a model to data, calibrating a simulation, training a machine learning model, or choosing engineering parameters that minimize cost. Basin gives users a single, consistent way to both state and solve such problems, with a broad catalog of solvers and first-class support for constraints.

Voltage certificates for droop-controlled radial distribution networks are often constructed from worst-case quantities. In heterogeneous radial networks, this approach can hide where voltage deviations are most likely. Furthermore, such certificates become increasingly conservative as the network sensitivities accumulate. Leveraging the structure of the linearized DistFlow model and slope-restricted droop controllers, we derive tighter deterministic voltage certificates that retain heterogeneous network, disturbance and inverter characteristics of each bus. The certificates reveal the buses and local limitations that dominate certified voltage performance. Worst-case bounds are recovered as a special case. Although tighter, the heterogeneous certificates can still deteriorate downstream due to the network structure. To address this, we leverage our certificates as design variables for coordinated voltage control. Specifically, we develop a virtual droop architecture with coordination and affine feedforward compensation to reshape the effective voltage sensitivity, and formulate a linear program that jointly synthesizes the controller and minimizes its heterogeneous voltage certificates under operational, communication and inverter placement constraints. The resulting controller guarantees the voltage and inverter bounds for all admissible disturbances. Evaluation on two network benchmarks, a five-customer residential feeder and a 26-customer rural network comprising four feeders, demonstrates improved certificates and voltage behaviour under inverter limits.

We propose a comprehensive framework for solving overlapping-generations (OLG) models in continuous time with both idiosyncratic and aggregate risk. Our general characterization of equilibrium through the master equation operates on the joint distribution over the continuous idiosyncratic states, age and wealth. Our computational strategy is to take a finite-dimensional representation of this distribution as an input of a neural net which in turn outputs a finite-difference representation of the (conditional) value function. This idea can be applied generally to heterogeneous agent models with aggregate risk, and we call it finite-difference neural operator. Our method combines advantages from modern neural nets and traditional finite-difference methods: It is grid-free in the high-dimensional distribution, and retains control on boundary conditions in low-dimensional state variables. Moreover, our method is able to enforce shape constraints. We showcase its flexibility by solving a continuous-time OLG model with aggregate risk alone where we characterize the distribution by its supporting function; and to an OLG model with both types of risk.

This paper addresses the integrated Master Surgical Scheduling Problem and Surgical Case Assignment Problem under weekend bed capacity constraints. We propose a joint optimization framework that simultaneously determines the assignment of specialties and surgeons to Operating Room (OR) blocks, the selection and assignment of patients to these blocks, and the sequencing of surgeries, while accounting for reduced bed availability during weekends. The model is formulated as a multi-criteria Mixed-Integer Linear Program (MILP) that prioritizes the total priority-weighted amount of scheduled patients, minimizes excess weekend bed occupancy, and maximizes OR utilization. To solve large instances, we develop a Column Generation (CG) algorithm in which the pricing subproblem is formulated as a Resource-Constrained Shortest Path Problem. The framework accommodates block, open, and modified block scheduling policies. Computational experiments show that the CG approach outperforms the direct MILP solution in terms of solution quality and computational time, particularly for larger instances. We further analyze the trade-offs between weekend bed availability, OR utilization, and total patient priority. Our results provide managerial insights into how hospitals can balance patient access, operational efficiency, and staff workload when resources are limited.

Clustering is a fundamental class of data analysis techniques with the most important representatives being centroid-based methods like $k$-means. Such methods are strongly connected to quantization problems, which aim to approximate general probability measures with discrete ones. For example, $k$-means corresponds to quantization with respect to the Wasserstein distance. While Wasserstein quantization clusters points within a fixed space, this paper studies Gromov-Wasserstein (GW) quantization, which additionally aims at clustering the ambient geometry of the space. We show existence of solutions to the GW quantization problem and give a characterization that justifies an analogue to the $k$-means algorithm (Lloyd's algorithm) to approximate them numerically. We further calculate the quantization rate for usual Euclidean geometries that are used in the GW context, and relate it to standard Wasserstein quantization rates. Finally, numerical experiments show that GW quantization opens up many modeling possibilities beyond normal clustering methods (e.g., for geodesic distances of 3D shapes or structured pruning of neural networks) and that the introduced algorithm leads to useful numerical solutions with approximation quality often in line with theoretically optimal rates.

Decision diagrams (DDs) have become a powerful tool for discrete optimization, supporting a wide range of algorithms that span cut-generation procedures, decomposition methods, and specialized branch-and-bound searches. Despite this growth, their adoption remains limited, partly because most existing DD code is tailored to a specific algorithm or application and is therefore hard to reuse. We introduce DD-suite, a cross-platform, open-source software package for building and manipulating DDs for discrete optimization. DD-suite is available in both Python and C++ through a shared modeling interface, and lets users construct exact, restricted, and relaxed DDs for any discrete optimization problem expressed in recursive form. The package implements the DD reduction procedure, shortest-path routines for obtaining primal and dual bounds, a visualization tool, and an extensive automated test suite. Furthermore, it includes extensive documentation, a support webpage, and ready-to-use examples for four combinatorial problems. Rather than a closed solver, DD-suite is designed as an extensible building block: users can add new construction mechanisms or run custom algorithms on top of the resulting diagram, as we illustrate with a DD-based cutting plane procedure embedded in a state-of-the-art mixed-integer programming solver. Our numerical experiments show that the C++ implementation is 5-6 times faster than the Python one while producing identical diagrams, and remains within a small constant factor of ddo, a specialized Rust framework, confirming that DD-suite combines an accessible, extensible codebase with competitive performance.

We study funnel control for linear infinite-dimensional passive systems whose controlled output is obtained by integrating the passive output. In mechanical applications, this corresponds to position control based on a measured passive velocity output. Within the system-node framework, we propose a nonlinear recursive funnel controller with a dynamic gain supervisor for systems with distributed or boundary control and observation. The supervisor automatically increases a damping gain whenever the normalized recursive error approaches a prescribed monitoring level. The controller uses the tracking error and its derivative, with the latter obtained directly from the passive output rather than by numerical differentiation. We prove global existence and uniqueness of the closed-loop trajectory, prescribed funnel performance, and global boundedness of the derivative tracking error and the passive output. Under an additional position-compatible strict dissipativity condition, the supervised gain is increased only finitely many times and eventually becomes constant. Consequently, the gain, the internal state, and the control input are globally bounded, and the recursive-error funnel margin is uniform in time. The results are illustrated by a boundary-actuated Euler-Bernoulli beam with uniformly positive distributed damping.

We study a Restart POMDP (Partially Observable Markov Decision Process) on a general Borel state space, where the controller either lets the hidden state evolve unobserved or restarts the system and observes the new state. Exploiting a sufficient-statistic representation consisting of the last observed state and the elapsed time since restart, we reduce the problem to a fully observed MDP. Under a natural one-step cost deterioration condition, we prove that optimal policies have a threshold structure in the elapsed time for both the discounted and total undiscounted cost criteria. When the state space is partially ordered and the kernel is stochastically monotone, we further show that the optimal threshold is nonincreasing in the state. For the average cost criterion, under additional assumptions of geometric ergodicity and domination of the transient gain, we establish analogous threshold results via the vanishing discount approach, after showing the uniform boundedness of the optimal thresholds and relative value functions.

We consider a patrol problem in which a patroller moves among several geographically dispersed locations to detect attackers that arrive over time. Once the patroller arrives at a location, they can spend any amount of time searching for attackers at that location—and detect an attacker there with some location-dependent instantaneous detection rate—before moving to a different location. The objective of the patroller is to minimize the expected time an attacker stays undetected at a location, regardless of where and when the attack occurs. In the special case where travel times are negligible, we elucidate an optimal cyclic policy in which the patroller allocates a fixed fraction of their effort to each location continuously. The patrol problem becomes significantly more challenging when travel times cannot be ignored. We introduce two types of cyclic patrol patterns based on common patrol practice for perimeter patrol and border patrol, respectively, and derive formulae for the expected time to detect an attack in both cases. We also provide an algorithm for finding the best search time parameters in both of these cases. We give several examples where these cycle types perform well and numerically demonstrate that the optimal patrol policy depends highly on the structure and parameters of each patrol problem.

While district heating networks are a key component of the energy transition, their operational planning is challenging due to substantial uncertainty in heat demand. We address this volatility without compromising system reliability. By adapting chance-constrained programming (CC) and conditional value-at-risk (CVaR) optimization to the mixed-integer unit-commitment problem for district-heating networks, we can compute optimal unit-commitment schedules under demand uncertainty. We generate heat demand time series using a Bayesian model, which explicitly quantifies forecasting uncertainty and yields predictive distributions as input to the network optimization. To handle the inherent uncertainty in heat demand, we apply both robust optimization approaches: chance-constrained programming limits the probability of unmet demand, while CVaR optimization penalizes severe shortfalls in the tail of the distribution. We evaluate the approaches both on real-world data from the Berlin district heating network and on a benchmark set of synthetic, realistically parameterized instances of varying sizes. The results compare the two uncertainty-handling methods with respect to solution quality, risk exposure, and computational effort.