arXiv subjects

Kimia Ghobadi

Publications and source records attributed to Kimia Ghobadi.

18 recordsLinked to original sources

arXiv2026

Joint Score-Threshold Optimization for Interpretable Risk Assessment

Risk assessment tools in healthcare commonly employ point-based scoring systems that map patients to ordinal risk categories via thresholds. While electronic health record (EHR) data presents opportunities for data-driven optimization of these tools, two fundamental challenges impede standard supervised learning: (1) labels are often available only for extreme risk categories due to intervention-censored outcomes, and (2) misclassification cost is asymmetric and increases with ordinal distance. We propose a mixed-integer programming (MIP) framework that jointly optimizes scoring weights and category thresholds in the face of these challenges. Our approach prevents label-scarce category collapse via threshold constraints, and utilizes an asymmetric, distance-aware objective. The MIP framework supports governance constraints, including sign restrictions, sparsity, and minimal modifications to incumbent tools, ensuring practical deployability in clinical workflows. We further develop a continuous relaxation of the MIP problem to provide warm-start solutions for more efficient MIP optimization. We apply the proposed score optimization framework to a case study of inpatient falls risk assessment using the Johns Hopkins Fall Risk Assessment Tool.

cs.LG↗

arXiv2026

Q-Learning with Shift-Aware Upper Confidence Bound in Non-Stationary Reinforcement Learning

We study the Non-Stationary Reinforcement Learning (RL) under distribution shifts in both finite-horizon episodic and infinite-horizon discounted Markov Decision Processes (MDPs). In the finite-horizon case, the transition functions may suddenly change at a particular episode. In the infinite-horizon setting, such changes can occur at an arbitrary time step during the agent's interaction with the environment. While the Q-learning Upper Confidence Bound algorithm (QUCB) can discover a proper policy during learning, due to the distribution shifts, this policy can exploit sub-optimal rewards after the shift happens. To address this issue, we propose Density-QUCB (DQUCB), a shift-aware Q-learning UCB algorithm, which uses a transition density function to detect distribution shifts, then leverages its likelihood to enhance the uncertainty estimation quality of Q-learning UCB, resulting in a balance between exploration and exploitation. Theoretically, we prove that our oracle DQUCB achieves a better regret guarantee than QUCB. Empirically, our DQUCB enjoys the computational efficiency of model-free RL and outperforms QUCB baselines by having a lower regret across RL tasks, as well as a COVID-19 patient hospital allocation task using a Deep-Q-learning architecture.

cs.LG↗

arXiv2026

From Non-Identifiability to Goal-Integrated Decision-Making in Parametric Inverse Optimization

Inverse optimization seeks to recover unknown objective parameters from observed decisions, yet fundamental questions about when recovery is possible have received limited formal treatment. This paper develops a comprehensive theoretical framework for inverse optimization in parametric convex models. We first establish that non-identifiability is the generic case: even with normalization and multiple observations, the parameter set compatible with data is generically multi-dimensional, and regularization does not resolve this. We derive necessary and sufficient conditions for identifiability. Motivated by these negative results, we introduce the Inverse Learning (IL) framework, which shifts the inferential target from the unknown parameter to the latent optimal solution, achieving a complexity reduction that is independent of the number of observations. IL explicitly characterizes the full set of compatible parameters rather than returning an arbitrary element. To address the tension between observational fidelity and constraint adherence, we formalize the Observation-Constraint Tradeoff and develop Goal-Integrated Inverse Learning models that enable structured navigation of this spectrum with guaranteed monotonicity. Numerical experiments demonstrate superior solution accuracy, higher parameter recovery rates, and significant computational speedups. We apply the framework to personalized dietary recommendations using NHANES data, proof-of-concept demonstrating improved glycemic control in a prospective feasibility study.

math.OC↗

arXiv2026

TsLLM: Augmenting LLMs for General Time Series Understanding and Prediction

Time series data is fundamental to decision-making across many domains including healthcare, finance, power systems, and logistics. However, analyzing this data correctly often requires incorporating unstructured contextual information, answering domain-specific questions, and generating natural language explanations - capabilities that traditional time series models lack. While Large Language Models (LLMs) excel at contextual reasoning and knowledge integration, they struggle with numerical time series due to inefficient text-based representations and limited exposure to numerical data during pretraining. We address this gap by augmenting an LLM with specialized time series perception through a patch-based encoder-decoder architecture. We train this Time Series augmented LLM (TsLLM) on a large corpus of over 25 billion tokens of interleaved time series and text spanning diverse tasks: forecasting with contextual information, question-answering, anomaly detection, classification, report generation, and more, all unified as next token prediction. This training enables TsLLM to leverage both its language understanding and newly acquired temporal reasoning capabilities. While not designed to surpass specialized models on traditional benchmarks, TsLLM demonstrates strong performance on tasks requiring the integration of time series analysis with natural language - capabilities that existing approaches cannot provide. It also exhibits strong zero-shot and few-shot performance, showing it can adapt to new data without additional training.

cs.LG↗

arXiv2026

Eliciting Chain-of-Thought Reasoning for Time Series Analysis using Reinforcement Learning

Complex numerical time series analysis often demands multi-step reasoning capabilities beyond current models' reach. Tasks like medical diagnosis and weather forecasting require sequential reasoning processes - including counterfactual analysis, logical deduction, knowledge application, and multi-modal contextual integration - that existing time series models cannot explicitly perform. While recent research has shown large language models (LLMs) can achieve sophisticated Chain-of-Thought (CoT) reasoning through reinforcement learning (RL), these advances have primarily focused on mathematical and coding domains, with LLMs still demonstrating poor performance on time series tasks. We introduce Chain Of thought for Understanding Numerical Time Series (COUNTS), the first framework that trains LLMs to perform CoT reasoning across diverse time series tasks using RL with verifiable rewards. Our approach employs a Residual Vector-Quantized VAE to create high-fidelity discrete tokens that seamlessly integrate into a pre-trained LLM's vocabulary. COUNTS undergoes a two-stage training process: first, supervised fine-tuning on time series analysis tasks to master our novel representations, followed by Group Relative Policy Optimization training on verifiable problems using prompting strategies that encourage explicit reasoning steps before producing final answers. Our experiments demonstrate that this RL-driven approach with intermediate CoT reasoning significantly enhances LLM performance across various time series analysis tasks, opening new possibilities for complex temporal data reasoning.

cs.LG↗

arXiv2026

An interpretable data-driven approach to optimizing clinical fall risk assessment

In this study we aim to better align fall risk prediction from the Johns Hopkins Fall Risk Assessment Tool (JHFRAT) with additional clinically meaningful measures via a data-driven modelling approach. We conducted a retrospective analysis of 54,209 inpatient admissions from three Johns Hopkins Health System hospitals between March 2022 and October 2023. A total of 20,208 admissions were included as high fall risk encounters, and 13,941 were included as low fall risk encounters. To incorporate clinical knowledge and maintain interpretability, we employed constrained score optimization (CSO) models on JHFRAT assessment data and additional electronic health record (EHR) variables. The model demonstrated significant improvements in predictive performance over the current JHFRAT (CSO AUC-ROC=0.91, JHFRAT AUC-ROC=0.86). The constrained score optimization models performed similarly with and without the EHR variables. Although the benchmark black-box model (XGBoost), improves upon the performance metrics of the knowledge-based constrained logistic regression (AUC-ROC=0.94), the CSO demonstrates more robustness to variations in risk labelling. This evidence-based approach provides a robust foundation for health systems to systematically enhance inpatient fall prevention protocols and patient safety using data-driven optimization techniques, contributing to improved risk assessment and resource allocation in healthcare settings.

cs.LG↗

arXiv2026

An interpretable data-driven approach to optimizing clinical fall risk assessment

In this study, we aim to better align fall risk prediction from the Johns Hopkins Fall Risk Assessment Tool (JHFRAT) with additional clinically meaningful measures via a data-driven modelling approach. We conducted a retrospective cohort analysis of 54,209 inpatient admissions from three Johns Hopkins Health System hospitals between March 2022 and October 2023. A total of 20,208 admissions were included as high fall risk encounters, and 13,941 were included as low fall risk encounters. To incorporate clinical knowledge and maintain interpretability, we employed constrained score optimization (CSO) models to reweight the JHFRAT scoring weights, while preserving its additive structure and clinical thresholds. Recalibration refers to adjusting item weights so that the resulting score can order encounters more consistently by the study's risk labels, and without changing the tool's form factor or deployment workflow. The model demonstrated significant improvements in predictive performance over the current JHFRAT (CSO AUC-ROC=0.91, JHFRAT AUC-ROC=0.86). This performance improvement translates to protecting an additional 35 high-risk patients per week across the Johns Hopkins Health System. The constrained score optimization models performed similarly with and without the EHR variables. Although the benchmark black-box model (XGBoost), improves upon the performance metrics of the knowledge-based constrained logistic regression (AUC-ROC=0.94), the CSO demonstrates more robustness to variations in risk labeling. This evidence-based approach provides a robust foundation for health systems to systematically enhance inpatient fall prevention protocols and patient safety using data-driven optimization techniques, contributing to improved risk assessment and resource allocation in healthcare settings.

cs.LG↗

arXiv2024

Expert-Guided Inverse Optimization for Convex Constraint Inference

Conventional inverse optimization inputs a solution and finds the parameters of an optimization model that render a given solution optimal. The literature mostly focuses on inferring the objective function in linear problems when accepted solutions are provided as input. In this paper, we propose an inverse optimization model that inputs several accepted and rejected solutions and recovers the underlying convex optimization model that can be used to generate such solutions. The novelty of our model is two-fold: First, we focus on inferring the parameters of the underlying convex feasible region. Second, the proposed model learns the convex constraint set from a set of past observations that are either accepted or rejected by an expert. The resulting inverse model is a mixed-integer nonlinear problem that is complex to solve. To mitigate the inverse problem complexity, we employ variational inequalities and the theoretical properties of the solutions to derive a reduced formulation that retains the complexity of its forward counterpart. Using realistic breast cancer patient data, we demonstrate that our inverse model can utilize a subset of past accepted and rejected treatment plans to infer clinical criteria that can lead to nearly guaranteed acceptable treatment plans for future patients.

math.OC↗

arXiv2024

MedTsLLM: Leveraging LLMs for Multimodal Medical Time Series Analysis

The complexity and heterogeneity of data in many real-world applications pose significant challenges for traditional machine learning and signal processing techniques. For instance, in medicine, effective analysis of diverse physiological signals is crucial for patient monitoring and clinical decision-making and yet highly challenging. We introduce MedTsLLM, a general multimodal large language model (LLM) framework that effectively integrates time series data and rich contextual information in the form of text to analyze physiological signals, performing three tasks with clinical relevance: semantic segmentation, boundary detection, and anomaly detection in time series. These critical tasks enable deeper analysis of physiological signals and can provide actionable insights for clinicians. We utilize a reprogramming layer to align embeddings of time series patches with a pretrained LLM's embedding space and make effective use of raw time series, in conjunction with textual context. Given the multivariate nature of medical datasets, we develop methods to handle multiple covariates. We additionally tailor the text prompt to include patient-specific information. Our model outperforms state-of-the-art baselines, including deep learning models, other LLMs, and clinical methods across multiple medical domains, specifically electrocardiograms and respiratory waveforms. MedTsLLM presents a promising step towards harnessing the power of LLMs for medical time series analysis that can elevate data-driven tools for clinicians and improve patient outcomes.

cs.LG↗

arXiv2024

An Interactive Decision-Support Dashboard for Optimal Hospital Capacity Management

Data-driven optimization models have the potential to significantly improve hospital capacity management, particularly during demand surges, when effective allocation of capacity is most critical and challenging. However, integrating models into existing processes in a way that provides value requires recognizing that hospital administrators are ultimately responsible for making capacity management decisions, and carefully building trustworthy and accessible tools for them. In this study, we develop an interactive, user-friendly, electronic dashboard for informing hospital capacity management decisions during surge periods. The dashboard integrates real-time hospital data, predictive analytics, and optimization models. It allows hospital administrators to interactively customize parameters, enabling them to explore a range of scenarios, and provides real-time updates on recommended optimal decisions. The dashboard was created through a participatory design process, involving hospital administrators in the development team to ensure practical utility, trustworthiness, transparency, explainability, and usability. We successfully deployed our dashboard within the Johns Hopkins Health System during the height of the COVID-19 pandemic, addressing the increased need for tools to inform hospital capacity management. It was used on a daily basis, with results regularly communicated to hospital leadership. This study demonstrates the practical application of a prospective, data-driven, interactive decision-support tool for hospital system capacity management.

cs.CY↗

arXiv2024

Improving Observed Decisions Quality using Inverse Optimization: A Radiation Therapy Treatment Planning Application

In many applied optimization settings, parameters that define the constraints may not guarantee the best possible solution, and superior solutions might exist that are infeasible for the given parameter values. Removing such constraints, re-optimizing, and evaluating the new solution may be insufficient, as the optimizer's preferences in selecting the existing solutions might be lost. To address this issue, we present an inverse optimization-based model that takes an observed solution as input and aims to improve upon it by projecting onto desired hyperplanes or expanding the feasible set while balancing the distance to the observed decision to preserve the optimizer's preferences. We demonstrate the applicability of the model in the context of radiation therapy treatment planning, an essential component of cancer treatment. Radiation therapy treatment planning is typically guided by expert-driven guidelines that define the optimization problem but remain mostly general. Our model provides an automated framework that learns new plans from available plans based on given clinical criteria, optimizing the desired effect without compromising the remaining constraints. The proposed approach is applied to a cohort of four prostate cancer patients, and the results demonstrate improvements in dose-volume histograms while maintaining comparable target coverage to clinically acceptable plans. By optimizing the parameters of the treatment planning problem and exploring the Pareto frontier, our methodology uncovers previously unattainable solutions that enhance organ-at-risk sparing without sacrificing target coverage. The framework's ability to handle multiple organs-at-risk and various dose-volume constraints highlights its flexibility and potential for application to diverse radiation therapy treatment planning scenarios.

math.OC↗

arXiv2024

Optimal Hospital Capacity Management During Demand Surges

Effective hospital capacity management is pivotal for enhancing patient care quality, operational efficiency, and healthcare system resilience, notably during demand spikes like those seen in the COVID-19 pandemic. However, devising optimal capacity strategies is complicated by fluctuating demand, conflicting objectives, and multifaceted practical constraints. This study presents a data-driven framework to optimize capacity management decisions within hospital systems during surge events. Two key decisions are optimized over a tactical planning horizon: allocating dedicated capacity to surge patients and transferring incoming patients between emergency departments (EDs) of hospitals to better distribute demand. The optimization models are formulated as robust mixed-integer linear programs, enabling efficient computation of optimal decisions that are robust against demand uncertainty. The models incorporate practical constraints and costs, including setup times and costs for adding surge capacity, restrictions on ED patient transfers, and relative costs of different decisions that reflect impacts on care quality and operational efficiency. The methodology is evaluated retrospectively in a hospital system during the height of the COVID-19 pandemic to demonstrate the potential impact of the recommended decisions. The results show that optimally allocating beds and transferring just 32 patients over a 63 day period around the peak, about one transfer every two days, could have reduced the need for surge capacity in the hospital system by nearly 90%. Overall, this work introduces a practical tool to transform capacity management decision-making, enabling proactive planning and the use of data-driven recommendations to improve outcomes.

cs.CY↗

arXiv2023

Inverse Learning: Solving Partially Known Models Using Inverse Optimization

We consider the problem of learning optimal solutions of a partially known linear optimization problem and recovering its underlying cost function where a set of past decisions and the feasible set are known. We develop a new framework, denoted as Inverse Learning, that extends the inverse optimization literature to (1) learn the optimal solution of the underlying problem, (2) integrate additional information on constraints and their importance, and (3) control the balance between mimicking past behaviors and reaching new goals and rules for the learned solution. We pose inverse learning as an optimization problem that maps given (feasible and infeasible) observations to a single optimal solution with minimum perturbation, hence, not only recovering the missing cost vector but also providing an optimal solution simultaneously. The framework provides insights into an essential tradeoff in recovering linear optimization problems with regard to preserving observed behaviors and binding constraints of the known feasible set at optimality. We propose a series of mixed integer linear programming models to capture the effects of this tradeoff and validate it using a two-dimensional example. We then demonstrate the framework's applicability to a diet recommendation problem for a population of hypertension and prediabetic patients. The goal is to balance dietary constraints to achieve the necessary nutritional goals with the dietary habits of the users to encourage adherence to the diet. Results indicate that our models recommend daily food intakes that preserve the original data trends while providing a range of options to patients and providers based on the tradeoff.

math.OC↗

arXiv2022

You Are What You Eat: A Preference-Aware Inverse Optimization Approach

A key challenge in the emerging field of precision nutrition entails providing diet recommendations that reflect both the (often unknown) dietary preferences of different patient groups and known dietary constraints specified by human experts. Motivated by this challenge, we develop a preference-aware constrained-inference approach in which the objective function of an optimization problem is not pre-specified and can differ across various segments. Among existing methods, clustering models from machine learning are not naturally suited for recovering the constrained optimization problems, whereas constrained inference models such as inverse optimization do not explicitly address non-homogeneity in given datasets. By harnessing the strengths of both clustering and inverse optimization techniques, we develop a novel approach that recovers the utility functions of a constrained optimization process across clusters while providing optimal diet recommendations as cluster representatives. Using a dataset of patients' daily food intakes, we show how our approach generalizes stand-alone clustering and inverse optimization approaches in terms of adherence to dietary guidelines and partitioning observations, respectively. The approach makes diet recommendations by incorporating both patient preferences and expert recommendations for healthier diets, leading to structural improvements in both patient partitioning and nutritional recommendations for each cluster. An appealing feature of our method is its ability to consider infeasible but informative observations for a given set of dietary constraints. The resulting recommendations correspond to a broader range of dietary options, even when they limit unhealthy choices.

math.OC↗

arXiv2020

Optimal Resource and Demand Redistribution for Healthcare Systems Under Stress from COVID-19

When facing an extreme stressor, such as the COVID-19 pandemic, healthcare systems typically respond reactively by creating surge capacity at facilities that are at or approaching their baseline capacity. However, creating individual capacity at each facility is not necessarily the optimal approach, and redistributing demand and critical resources between facilities can reduce the total required capacity. Data shows that this additional load was unevenly distributed between hospitals during the COVID-19 pandemic, requiring some to create surge capacity while nearby hospitals had unused capacity. Not only is this inefficient, but it also could lead to a decreased quality of care at over-capacity hospitals. In this work, we study the problem of finding optimal demand and resource transfers to minimize the required surge capacity and resource shortage during a period of heightened demand. We develop and analyze a series of linear and mixed-integer programming models that solve variants of the demand and resource redistribution problem. We additionally consider demand uncertainty and use robust optimization to ensure solution feasibility. We also incorporate a range of operational constraints and costs that decision-makers may need to consider when implementing such a scheme. Our models are validated retrospectively using COVID-19 hospitalization data from New Jersey, Texas, and Miami, yielding at least an 85% reduction in required surge capacity relative to the observed outcome of each case. Results show that such solutions are operationally feasible and sufficiently robust against demand uncertainty. In summary, this work provides decision-makers in healthcare systems with a practical and flexible tool to reduce the surge capacity necessary to properly care for patients in cases when some facilities are over capacity.

math.OC↗

arXiv2020

An Open-Source Dataset on Dietary Behaviors and DASH Eating Plan Optimization Constraints

Linear constrained optimization techniques have been applied to many real-world settings. In recent years, inferring the unknown parameters and functions inside an optimization model has also gained traction. This inference is often based on existing observations and/or known parameters. Consequently, such models require reliable, easily accessed, and easily interpreted examples to be evaluated. To facilitate research in such directions, we provide a modified dataset based on dietary behaviors of different groups of people, their demographics, and pre-existing conditions, among other factors. This data is gathered from the National Health and Nutrition Examination Survey (NHANES) and complemented with the nutritional data from the United States Department of Agriculture (USDA). We additionally provide tailored datasets for hypertension and pre-diabetic patients as groups of interest who may benefit from targetted diets such as the Dietary Approaches to Stop Hypertension (DASH) eating plan. The data is compiled and curated in such a way that it is suitable as input to linear optimization models. We hope that this data and its supplementary, open-accessed materials can accelerate and simplify interpretations and research on linear optimization and constrained inference models. The complete dataset can be found in the following repository: https://github.com/CSSEHealthcare/InverseLearning

math.OC↗

arXiv2020

Inferring Linear Feasible Regions using Inverse Optimization

Consider a problem where a set of feasible observations are provided by an expert and a cost function is defined that characterizes which of the observations dominate the others and are hence, preferred. Our goal is to find a set of linear constraints that would render all the given observations feasible while making the preferred ones optimal for the cost (objective) function. By doing so, we infer the implicit feasible region of the linear programming problem. Providing such feasible regions (i) builds a baseline for categorizing future observations as feasible or infeasible, and (ii) allows for using sensitivity analysis to discern changes in optimal solutions if the objective function changes in the future. In this paper, we propose an inverse optimization framework to recover the constraints of a forward optimization problem using multiple past observations as input. We focus on linear models in which the objective function is known but the constraint matrix is partially or fully unknown. We propose a general inverse optimization methodology that recovers the complete constraint matrix and then introduce a tractable equivalent reformulation. Furthermore, we provide and discuss several generalized loss functions to inform the desirable properties of the feasible region based on user preference and historical data. Our numerical examples verify the validity of our approach, emphasize the differences among the proposed measures, and provide intuition for large-scale implementations. We further demonstrate our approach using a diet recommendation problem to show how the proposed models can help impute personalized constraints for each dieter.

math.OC↗

arXiv2020

Artificial Intelligence-based Clinical Decision Support for COVID-19 -- Where Art Thou?

The COVID-19 crisis has brought about new clinical questions, new workflows, and accelerated distributed healthcare needs. While artificial intelligence (AI)-based clinical decision support seemed to have matured, the application of AI-based tools for COVID-19 has been limited to date. In this perspective piece, we identify opportunities and requirements for AI-based clinical decision support systems and highlight challenges that impact "AI readiness" for rapidly emergent healthcare challenges.

cs.CY↗