Sheryl Paul

Networked multi-agent dynamical systems have been used to model how individual opinions evolve over time due to the opinions of other agents in the network. Particularly, such a model has been used to study how a planning agent can be used to steer opinions in a desired direction through repeated, budgeted interventions. In this paper, we consider the problem where individuals' susceptibilities to external influences are unknown.

We propose an online algorithm that alternates between estimating this susceptibility parameter, and using the current estimate to drive the opinion to a desired target. We provide conditions that guarantee stability and convergence to the desired target opinion when the planning agent faces budgetary or temporal constraints. Our analysis shows that the key advantage of estimating the susceptibility parameter is that it helps achieve near-optimal convergence to the target opinion given a finite amount of intervention rounds, and, for a given intervention budget, quantifies how close the opinion can get to the desired target.

Multi-agent planning problems arise in a variety of engineering applications, such as multi-robot wildfire fighting and unmanned aerial inspection in factories. A particular challenge is the existence of spatio-temporal (i.e., when an agent should do what) and topological constraints (i.e., how agents should interact), as typically formalized via the notion of graphs. We focus here on spatio-temporal logic with graph operators (STL-GO), a recent formalism that allows to reason over multiple agents and their topologies, such as sensing, communication, and task topologies.

In this paper, we consider the problem of planning multi-agent paths that satisfy constraints written in STL-GO. We present two solutions to this problem based on mixed integer programming (MIP) and satisfiability modulo theory (SMT) solving with soundness guarantees. We provide a unified interface for specifying agents, their graph topologies, and the STL-GO specification, enabling seamless use of both methods and facilitating direct comparison between them. We evaluate both encodings on a multi-UAV search-and-rescue benchmark, ablating over team size and graph complexity highlighting the expressiveness of the proposed encodings under dynamic multi-graph interactions.

Spatio-Temporal Logic with Graph Operators (STL-GO) extends Signal Temporal Logic (STL) to multi-agent systems by introducing graph operators that reason about the number of neighboring agents satisfying a property, and multi-agent quantifiers. While Boolean semantics for STL-GO are well-defined, quantitative semantics have not been developed. Existing quantitative semantics for spatio-temporal logics such as STREL are insufficient for the counting constraints in STL-GO's graph operators.

Thus, we develop quantitative semantics for STL-GO, as a layered algebraic construction that separates temporal aggregation from graph-operator aggregation (governed by an abstract accumulator with a monotone fold and readout). We prove that soundness and completeness of the full semantics reduce to monotonicity conditions on these components, and show that standard min-max and Boolean instantiations satisfy them automatically. We implement the framework and evaluate it on two multi-agent environments: a 2D bounded region with stochastic Dubins-car dynamics and a 3D Earth-satellite system, under four semantic instantiations, demonstrating the tradeoffs between accumulator choices and reporting scalability in the number of agents and time horizon.

Modern autonomous planning methods rely heavily on differentiable predictive models of agent and world dynamics, enabling gradient-based optimization through neural pipelines. In parallel, temporal logic formalisms provide expressive and unambiguous symbolic specifications of high-level task objectives. However, there is work to be done in unifying differentiable dynamics with expressive multi-agent spatio-temporal specifications, particularly in settings where the underlying interaction graph evolves with agent states.

Hence, we present a neuro-symbolic framework for multi-agent planning under Spatio-Temporal Reach and Escape Logic (STREL) specifications. We introduce smooth adjacency graph constructions that enable differentiation through dynamic interaction topologies, and propose STREL-Net, a Graph Neural Network (GNN)-based exact encoding of STREL quantitative semantics for gradient-based trajectory optimization. We evaluate our method in a customized SwarmLab environment and compare against SMT- and MILP-based STREL synthesis approaches across multiple specifications, linear and non-linear dynamics, varying team sizes, and settings with and without quadratic optimization objectives. Across all configurations, our approach consistently achieves substantially lower runtimes and remains feasible even in regimes where MILP and SMT formulations either become intractable or fail to solve.

We consider the multi-agent path planning problem in dynamically changing environments under spatio-temporal specifications. Each agent operates under partial observability, perceiving only the local state of the environment. To address this challenge, we propose a centralized planning framework that takes in each agent's observations of the world. Spatio-temporal objectives are encoded as constraints over trajectories of the agents. We formulate path planning as a dynamic programming problem over joint observation space, and generate plans that maximize the probability of satisfying the given spatio-temporal specification. We demonstrate the effectiveness of our approach in a customized 3D grid wildfire simulator.

Congestion pricing is a widely used tool for improving efficiency in transportation networks, where decentralized route choices by individual users can lead to significant system-level inefficiencies. A classical solution is Pigouvian pricing, which achieves the system optimum. However, it is inherently demand-dependent and can lose effectiveness under demand shifts, particularly on structurally critical bottleneck edges.

We propose a topology-aware regularization of the Beckmann potential, whose minimizer characterizes the Wardrop equilibrium induced by selfish routing. Our approach augments precomputed Pigouvian tolls with a structural penalty derived from Forman–Ricci curvature (FRC), which captures bottleneck structure and can be computed offline. We show that the resulting equilibrium is well-defined, and that increasing the regularization parameter reduces aggregate flow on penalized edges. We further show that on modular networks, FRC assigns the largest penalties to bridge edges that all inter-region traffic must traverse. Experiments on eight real-world networks under targeted and uniform demand perturbations show that the method reduces bottleneck flow while maintaining near-optimal efficiency. Compared to edge betweenness, FRC achieves comparable bottleneck reduction at substantially lower computational cost.

Existing models of mixed-autonomy weaving ramps characterize how altruistic connected and automated vehicles (CAVs) can improve traffic efficiency at the population level, but provide limited insight into how such behavior emerges from decentralized vehicle interactions or how it is affected by finite populations, heterogeneous preferences, and imperfect information. We develop an agent-based model of a macroscopic weaving-ramp framework in which individual vehicles adapt their lane choices using an evolutionary game-theoretic update rule and altruism-based objectives, providing a microscopic interpretation of the original Wardrop model. We prove convergence of the decentralized dynamics to the unique equilibrium predicted by the macroscopic theory.

Beyond reproducing aggregate equilibrium behavior, the framework enables the study of deployment-level questions that cannot be addressed by static analysis. Simulation results demonstrate close agreement with the macroscopic predictions while revealing how convergence rates, adaptation to changing traffic conditions, heterogeneous altruism levels among CAVs, and imperfect state information influence system performance and the distribution of altruistic burden across vehicles. These results provide a bridge between equilibrium traffic theory and decentralized mixed-autonomy deployment.

Modern cyber-physical systems (CPS) can consist of various networked components and agents interacting and communicating with each other. In the context of spatially distributed CPS, these connections can be dynamically dependent on the spatial configuration of the various components and agents. In these settings, robust monitoring of the distributed components is vital to ensuring complex behaviors are achieved, and safety properties are maintained.

To this end, we look at defining the automaton semantics for the Spatio-Temporal Reach and Escape Logic (STREL), a formal logic designed to express and monitor spatio-temporal requirements over mobile, spatially distributed CPS. Specifically, STREL reasons about spatio-temporal behavior over dynamic weighted graphs. While STREL is endowed with well-defined qualitative and quantitative semantics, in this paper, we propose a novel construction of an alternating finite automata-based monitor for STREL specifications.

Reinforcement learning (RL) has been successfully applied to solve the problem of finding obstacle-free paths for autonomous agents operating in stochastic and uncertain environments. However, when the underlying stochastic dynamics of the environment experiences drastic distribution shifts, the optimal policy obtained in the trained environment may be sub-optimal or may entirely fail in helping find goal-reaching paths for the agent. Approaches like domain randomization and robust RL can provide robust policies, but typically assume minor (bounded) distribution shifts.

For substantial distribution shifts, retraining is an alternative approach. In this paper, we develop a novel approach called Evolutionary Robust Policy Optimization (ERPO), an adaptive re-training algorithm inspired by evolutionary game theory (EGT). ERPO learns an optimal policy for the shifted environment iteratively using a temperature parameter that controls the trade off between exploration and adherence to the old optimal policy. The policy update itself is an instantiation of the replicator dynamics used in EGT. We show that under fairly common sparsity assumptions on rewards, ERPO converges to the optimal policy in the shifted environment, and empirically outperforms popular RL and deep RL algorithms (PPO, A3C, DQN) across many path-finding scenarios.

Autonomous multi-agent systems such as hospital robots and package delivery drones often operate in highly uncertain environments and are expected to achieve complex temporal task objectives while ensuring safety. While learning-based methods such as reinforcement learning are popular for training single and multi-agent autonomous systems under user-specified and state-based reward functions, applying these methods to satisfy trajectory-level task objectives is a challenging problem.

Our first contribution is the use of weighted automata to specify trajectory-level objectives, such that maximal paths induced in the weighted automaton correspond to desired trajectory-level behaviors. We show how weighted automata-based specifications go beyond timeliness properties focused on deadlines to performance properties such as expeditiousness. Our second contribution is the use of evolutionary game theory as a multi-agent path planning mechanism that produces trajectories satisfying the weighted automaton specification.

Natural language is an intuitive way for humans to communicate formal requirements of cyber-physical systems, such as safety specifications, performance requirements, and task objectives with autonomous CPS such as robots. While natural language (NL) is ambiguous, real world tasks and their safety requirements need to be communicated unambiguously. Signal Temporal Logic (STL) is a formal logic that can serve as a versatile, expressive, and unambiguous formal language to describe robotic tasks.

On one hand, existing work in using STL for the robotics domain typically requires end-users to express task specifications in STL, which is a challenge for non-expert users. On the other, translating from NL to STL specifications is currently restricted to specific fragments. In this work, we propose DialogueSTL, an explainable and interactive approach for learning correct and concise STL specifications from natural language descriptions.

Posing a threat to a large portion of the population on a worldwide scale, breast cancer is now a deleterious pandemic. Chances of survival greatly improve if detected early, giving us impetus to contribute to various methods of detection from a technical perspective. A tool used to detect and diagnose breast-related diseases is a digital mammogram. The main purpose of this paper is to detect the probability of early stage breast cancer detection using mammography images.

These mammograms were pre-processed using the CLAHE technique. In order to detect masses from the aforementioned mammograms, a hybrid approach has been employed: combining a neural network with a linear classifier. Deep Learning model VGG16, based on convolutional neural networks, was used for feature extraction, which were then fed into linear classifiers.