arXiv subjects

Shiqiang Zhang

Publications and source records attributed to Shiqiang Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Global Optimization of Gaussian Process Acquisition Functions Using a Piecewise-Linear Kernel Approximation

Bayesian optimization relies on iteratively constructing and optimizing an acquisition function. The latter turns out to be a challenging, non-convex optimization problem itself. Despite the relative importance of this step, most algorithms employ sampling- or gradient-based methods, which do not provably converge to global optima. This work investigates mixed-integer programming (MIP) as a paradigm for global acquisition function optimization. Specifically, our Piecewise-linear Kernel Mixed Integer Quadratic Programming (PK-MIQP) formulation introduces a piecewise-linear approximation for Gaussian process kernels and admits a corresponding MIQP representation for acquisition functions. The proposed method is applicable to uncertainty-based acquisition functions for any stationary or dot-product kernel. We analyze the theoretical regret bounds of the proposed approximation, and empirically demonstrate the framework on synthetic functions, constrained benchmarks, and a hyperparameter tuning task.

math.OC↗

arXiv2026

BoGrape: Bayesian optimization over graphs with shortest-path encoded

Graph-structured data are central to many scientific and industrial applications where the goal is to optimize expensive black-box objectives defined over graph structures or node configurations -- as seen in molecular design, supply chains, and sensor placement. Bayesian optimization offers a principled approach for such settings, but existing methods largely focus on functions defined over nodes of a fixed graph. Moreover, graph optimization is often approached heuristically, and it remains unclear how to systematically incorporate structural constraints into BO. To address these gaps, we build on shortest-path graph kernels to develop a principled framework for acquisition optimization over unseen graph structures and associated node attributes. Through a novel formulation based on mixed-integer programming, we enable global exploration of the combinatorial domain over graph structures and explicit embedding of problem-specific constraints. We demonstrate that our method, BoGrape, is competitive both on general synthetic benchmarks and representative molecular design case studies with application-specific constraints.

math.OC↗

arXiv2025

The Catechol Benchmark: Time-series Solvent Selection Data for Few-shot Machine Learning

Machine learning has promised to change the landscape of laboratory chemistry, with impressive results in molecular property prediction and reaction retro-synthesis. However, chemical datasets are often inaccessible to the machine learning community as they tend to require cleaning, thorough understanding of the chemistry, or are simply not available. In this paper, we introduce a novel dataset for yield prediction, providing the first-ever transient flow dataset for machine learning benchmarking, covering over 1200 process conditions. While previous datasets focus on discrete parameters, our experimental set-up allow us to sample a large number of continuous process conditions, generating new challenges for machine learning models. We focus on solvent selection, a task that is particularly difficult to model theoretically and therefore ripe for machine learning applications. We showcase benchmarking for regression algorithms, transfer-learning approaches, feature engineering, and active learning, with important applications towards solvent replacement and sustainable manufacturing.

cs.LG↗

arXiv2025

Beyond adjacency: Graph encoding with reachability and shortest paths

Graph-structured data is central to many scientific and industrial domains, where the goal is often to optimize objectives defined over graph structures. Given the combinatorial complexity of graph spaces, such optimization problems are typically addressed using heuristic methods, and it remains unclear how to systematically incorporate structural constraints to effectively reduce the search space. This paper introduces explicit optimization formulations for graph search space that encode properties such as reachability and shortest paths. We provide theoretical guarantees demonstrating the correctness and completeness of our graph encoding. To address the symmetry issues arising from graph isomorphism, we propose lexicographic constraints over neighborhoods to eliminate symmetries and theoretically prove that adding those constraints will not reduce the original graph space. Our graph encoding, along with the corresponding symmetry-breaking constraints, forms the basis for downstream optimization tasks over graph spaces.

math.OC↗

arXiv2025

Global optimization of graph acquisition functions for neural architecture search

Graph Bayesian optimization (BO) has shown potential as a powerful and data-efficient tool for neural architecture search (NAS). Most existing graph BO works focus on developing graph surrogates models, i.e., metrics of networks and/or different kernels to quantify the similarity between networks. However, the acquisition optimization, as a discrete optimization task over graph structures, is not well studied due to the complexity of formulating the graph search space and acquisition functions. This paper presents explicit optimization formulations for graph input space including properties such as reachability and shortest paths, which are used later to formulate graph kernels and the acquisition function. We theoretically prove that the proposed encoding is an equivalent representation of the graph space and provide restrictions for the NAS domain with either node or edge labels. Numerical results over several NAS benchmarks show that our method efficiently finds the optimal architecture for most cases, highlighting its efficacy.

cs.LG↗

arXiv2024

Limeade: Let integer molecular encoding aid

Mixed-integer programming (MIP) is a well-established framework for computer-aided molecular design (CAMD). By precisely encoding the molecular space and score functions, e.g., a graph neural network, the molecular design problem is represented and solved as an optimization problem, the solution of which corresponds to a molecule with optimal score. However, both the extremely large search space and complicated scoring process limit the use of MIP-based CAMD to specific and tiny problems. Moreover, optimal molecule may not be meaningful in practice if scores are imperfect. Instead of pursuing optimality, this paper exploits the ability of MIP in molecular generation and proposes Limeade as an end-to-end tool from real-world needs to feasible molecules. Beyond the basic constraints for structural feasibility, Limeade supports inclusion and exclusion of SMARTS patterns, automating the process of interpreting and formulating chemical requirements to mathematical constraints.

cs.CE↗

arXiv2024

Verifying message-passing neural networks via topology-based bounds tightening

Since graph neural networks (GNNs) are often vulnerable to attack, we need to know when we can trust them. We develop a computationally effective approach towards providing robust certificates for message-passing neural networks (MPNNs) using a Rectified Linear Unit (ReLU) activation function. Because our work builds on mixed-integer optimization, it encodes a wide variety of subproblems, for example it admits (i) both adding and removing edges, (ii) both global and local budgets, and (iii) both topological perturbations and feature modifications. Our key technology, topology-based bounds tightening, uses graph structure to tighten bounds. We also experiment with aggressive bounds tightening to dynamically change the optimization constraints by tightening variable bounds. To demonstrate the effectiveness of these strategies, we implement an extension to the open-source branch-and-cut solver SCIP. We test on both node and graph classification problems and consider topological attacks that both add and remove edges.

math.OC↗

arXiv2023

Augmenting optimization-based molecular design with graph neural networks

Computer-aided molecular design (CAMD) studies quantitative structure-property relationships and discovers desired molecules using optimization algorithms. With the emergence of machine learning models, CAMD score functions may be replaced by various surrogates to automatically learn the structure-property relationships. Due to their outstanding performance on graph domains, graph neural networks (GNNs) have recently appeared frequently in CAMD. But using GNNs introduces new optimization challenges. This paper formulates GNNs using mixed-integer programming and then integrates this GNN formulation into the optimization and machine learning toolkit OMLT. To characterize and formulate molecules, we inherit the well-established mixed-integer optimization formulation for CAMD and propose symmetry-breaking constraints to remove symmetric solutions caused by graph isomorphism. In two case studies, we investigate fragment-based odorant molecular design with more practical requirements to test the compatibility and performance of our approaches.

cs.CE↗

arXiv2023

Practical Path-based Bayesian Optimization

There has been a surge in interest in data-driven experimental design with applications to chemical engineering and drug manufacturing. Bayesian optimization (BO) has proven to be adaptable to such cases, since we can model the reactions of interest as expensive black-box functions. Sometimes, the cost of this black-box functions can be separated into two parts: (a) the cost of the experiment itself, and (b) the cost of changing the input parameters. In this short paper, we extend the SnAKe algorithm to deal with both types of costs simultaneously. We further propose extensions to the case of a maximum allowable input change, as well as to the multi-objective setting.

cs.LG↗

arXiv2023

Edge-assisted U-Shaped Split Federated Learning with Privacy-preserving for Internet of Things

In the realm of the Internet of Things (IoT), deploying deep learning models to process data generated or collected by IoT devices is a critical challenge. However, direct data transmission can cause network congestion and inefficient execution, given that IoT devices typically lack computation and communication capabilities. Centralized data processing in data centers is also no longer feasible due to concerns over data privacy and security. To address these challenges, we present an innovative Edge-assisted U-Shaped Split Federated Learning (EUSFL) framework, which harnesses the high-performance capabilities of edge servers to assist IoT devices in model training and optimization process. In this framework, we leverage Federated Learning (FL) to enable data holders to collaboratively train models without sharing their data, thereby enhancing data privacy protection by transmitting only model parameters. Additionally, inspired by Split Learning (SL), we split the neural network into three parts using U-shaped splitting for local training on IoT devices. By exploiting the greater computation capability of edge servers, our framework effectively reduces overall training time and allows IoT devices with varying capabilities to perform training tasks efficiently. Furthermore, we proposed a novel noise mechanism called LabelDP to ensure that data features and labels can securely resist reconstruction attacks, eliminating the risk of privacy leakage. Our theoretical analysis and experimental results demonstrate that EUSFL can be integrated with various aggregation algorithms, maintaining good performance across different computing capabilities of IoT devices, and significantly reducing training time and local computation overhead.

cs.LG↗

arXiv2023

Optimizing over trained GNNs via symmetry breaking

Optimization over trained machine learning models has applications including: verification, minimizing neural acquisition functions, and integrating a trained surrogate into a larger decision-making problem. This paper formulates and solves optimization problems constrained by trained graph neural networks (GNNs). To circumvent the symmetry issue caused by graph isomorphism, we propose two types of symmetry-breaking constraints: one indexing a node 0 and one indexing the remaining nodes by lexicographically ordering their neighbor sets. To guarantee that adding these constraints will not remove all symmetric solutions, we construct a graph indexing algorithm and prove that the resulting graph indexing satisfies the proposed symmetry-breaking constraints. For the classical GNN architectures considered in this paper, optimizing over a GNN with a fixed graph is equivalent to optimizing over a dense neural network. Thus, we study the case where the input graph is not fixed, implying that each edge is a decision variable, and develop two mixed-integer optimization formulations. To test our symmetry-breaking strategies and optimization formulations, we consider an application in molecular design.

math.OC↗

arXiv2023

SnAKe: Bayesian Optimization with Pathwise Exploration

Bayesian Optimization is a very effective tool for optimizing expensive black-box functions. Inspired by applications developing and characterizing reaction chemistry using droplet microfluidic reactors, we consider a novel setting where the expense of evaluating the function can increase significantly when making large input changes between iterations. We further assume we are working asynchronously, meaning we have to select new queries before evaluating previous experiments. This paper investigates the problem and introduces 'Sequential Bayesian Optimization via Adaptive Connecting Samples' (SnAKe), which provides a solution by considering large batches of queries and preemptively building optimization paths that minimize input costs. We investigate some convergence properties and empirically show that the algorithm is able to achieve regret similar to classical Bayesian Optimization algorithms in both synchronous and asynchronous settings, while reducing input costs significantly. We show the method is robust to the choice of its single hyper-parameter and provide a parameter-free alternative.

cs.LG↗

arXiv2021

Geometric Characteristics of Wasserstein Metric on SPD(n)

Wasserstein distance, especially among symmetric positive-definite matrices, has broad and deep influences on development of artificial intelligence (AI) and other branches of computer science. A natural idea is to describe the geometry of $SPD\left(n\right)$ as a Riemannian manifold endowed with the Wasserstein metric. In this paper, by involving the fiber bundle, we obtain explicit expressions for some locally geometric quantities, including geodesics, exponential maps, the Riemannian connection, Jacobi fields and curvatures. Furthermore, we discuss the behaviour of geodesics and prove that the manifold is globally geodesic convex with non-negative curvatures but no conjugate pair and cut locus. According to arithmetic estimates, we find curvatures can be controlled by the minimal eigenvalue.

math.DG↗

arXiv2020

Efficient Weingarten Map and Curvature Estimation on Manifolds

In this paper, we propose an efficient method to estimate the Weingarten map for point cloud data sampled from manifold embedded in Euclidean space. A statistical model is established to analyze the asymptotic property of the estimator. In particular, we show the convergence rate as the sample size tends to infinity. We verify the convergence rate through simulated data and apply the estimated Weingarten map to curvature estimation and point cloud simplification to multiple real data sets.

stat.ML↗

arXiv2020

The edge metric dimension of the generalized Petersen graph $P(n,3)$ is 4

It is known that the problem of computing the edge dimension of a graph is NP-hard, and that the edge dimension of any generalized Petersen graph $P(n,k)$ is at least 3. We prove that the graph $P(n,3)$ has edge dimension 4 for $n\ge 11$, by showing semi-combinatorially the nonexistence of an edge resolving set of order 3 and by constructing explicitly an edge resolving set of order 4.

math.CO↗

arXiv2020

Torsion Discriminance for Stability of Linear Time-Invariant Systems

This paper proposes a new approach to describe the stability of linear time-invariant systems via the torsion $τ(t)$ of the state trajectory. For a system $\dot{r}(t)=Ar(t)$ where $A$ is invertible, we show that (1) if there exists a measurable set $E_1$ with positive Lebesgue measure, such that $r(0)\in E_1$ implies that $\lim\limits_{t\to+\infty}τ(t)\neq0$ or $\lim\limits_{t\to+\infty}τ(t)$ does not exist, then the zero solution of the system is stable; (2) if there exists a measurable set $E_2$ with positive Lebesgue measure, such that $r(0)\in E_2$ implies that $\lim\limits_{t\to+\infty}τ(t)=+\infty$, then the zero solution of the system is asymptotically stable. Furthermore, we establish a relationship between the $i$th curvature $(i=1,2,\cdots)$ of the trajectory and the stability of the zero solution when $A$ is similar to a real diagonal matrix.

math.OC↗

arXiv2019

Self-organized patterns by a DC pin liquid anode discharge in ambient air: Effect of liquid types on formation

A pin liquid anode DC discharge is generated in open air without any additional gas feeding to form self-organized patterns (SOPs) on various liquid interfaces. Axially resolved emission spectra of the whole discharge reveal that the self-organized patterns are formed below a dark region and are visible mainly due to the N2 transitions. The high energy N2 (C) level is mainly excited by the impact of electrons heated by the local increased electric field at the interface. For the first time, the effect of the liquid type on the SOP formation is presented. With almost the same other discharge conditions, the formed SOPs are significantly different from HCl and H2SO4 liquid anodes. The SOP difference is repeated when the discharge current and gap distance change for both liquid anodes. The variations of SOP size and discretization as a function of discharge current and gap distance are discussed and confirm that different SOPs are formed by the HCl liquid anode from tap water or the H2SO4 liquid anode. A possible explanation is brought up to explain the dependence of SOPs on the liquid type.

physics.plasm-ph↗

arXiv2018

Description of Stability for Two and Three-Dimensional Linear Time-Invariant Systems Based on Curvature and Torsion

This paper focuses on using curvature and torsion to describe the stability of linear time-invariant system. We prove that for a two-dimensional system $\dot{r}(t)= Ar(t)$, (i) if there exists an initial value, such that zero is not the limit of curvature of trajectory as $t\to+\infty$, then the zero solution of the system is stable; (ii) if there exists an initial value, such that the limit of curvature of trajectory is infinity as $t\to+\infty$, then the zero solution of the system is asymptotically stable. For a three-dimensional system, (i) if there exists a measurable set whose Lebesgue measure is greater than zero, such that for all initial values in this set, zero is not the limit of curvature of trajectory as $t\to+\infty$, then the zero solution of the system is stable; (ii) if the coefficient matrix is invertible, and there exists a measurable set whose Lebesgue measure is greater than zero, such that for all initial values in this set, the limit of curvature of trajectory is infinity as $t\to+\infty$, then the zero solution of the system is asymptotically stable; (iii) if there exists a measurable set whose Lebesgue measure is greater than zero, such that for all initial values in this set, zero is not the limit of torsion of trajectory as $t\to+\infty$, then the zero solution of the system is asymptotically stable.

math.OC↗