Search for: All records

Generative artificial intelligence (GenAI) is shifting from conversational assistants toward agentic systems—autonomous decision-making systems that sense, decide, and act within operational workflows. This shift creates an autonomy paradox: as GenAI systems are granted greater operational autonomy, they should, by design, embody more formal structure, more explicit constraints, and stronger tail-risk discipline. We argue that stochastic generative models can be fragile in operational domains unless paired with mechanisms that provide verifiable feasibility, robustness to distribution shift, and stress testing under high-consequence scenarios. To address this challenge, we develop a conceptual framework for assured autonomy grounded in operations research (OR), built on two complementary approaches. First,flow-based generative modelsframe generation as deterministic transport characterized by an ordinary differential equation, enabling auditability, constraint-aware generation, and connections to optimal transport, robust optimization, and sequential decision control. Second,operational safetyis formulated through an adversarial robustness lens: decision rules are evaluated against worst-case perturbations within uncertainty or ambiguity sets, making unmodeled risks part of the design. This framework clarifies how increasing autonomy shifts OR’s role from solver to guardrail to system architect, with responsibility for control logic, incentive protocols, monitoring regimes, and safety boundaries. These elements define a research agenda forassured autonomyin safety-critical, reliability-sensitive operational domains. 

Entropy-Regularized Wasserstein Distributionally Robust Optimization Uncertainty in data poses a central challenge in operations research. Distributionally robust optimization (DRO) offers a principled framework for addressing this challenge by producing solutions resilient to distributional variations. Among various DRO approaches, the Wasserstein DRO has received significant attention though its computational efficiency relies on stringent assumptions, and its worst case distributions are typically discrete. In “Sinkhorn Distributionally Robust Optimization,” Wang, Gao, and Xie leverage the Sinkhorn distance—an entropy-regularized variant of the Wasserstein distance—to more realistically model uncertainty, enhancing computational efficiency. The authors establish a strong duality reformulation and propose a first order stochastic mirror descent algorithm with provable complexity guarantees for general loss functions. Unlike Wasserstein DRO, Sinkhorn DRO yields continuous worst case distributions, offering a more flexible representation of practical uncertainties. Extensive experiments in the newsvendor problem, portfolio optimization, and adversarial classification demonstrate its superior performance in both out-of-sample performance and efficiency. 

Flow Matching (FM) is a simulation-free method for learning a continuous, invertible flow that interpolates between two distributions, and in particular generates data from noise. Inspired by the variational nature of the diffusion process as a gradient flow, we introduce a stepwise FM model, Local Flow Matching (LFM), which sequentially learns a sequence of FM submodels, each matching a diffusion process up to the time-step size in the data-to-noise direction. In each step, the two distributions to be interpolated by the sub-flow model are closer than those in the full-flow matching model, which interpolates data to noise distributions, enabling smaller models with more efficient training. This variational perspective also allows us to prove a theoretical generation guarantee for the proposed flow model in terms of the χ2-divergence between the generated and true data distributions, leveraging the contraction property of the diffusion process. In practice, the stepwise structure of LFM is naturally amenable to model distillation, and various distillation techniques can be applied to accelerate generation. We empirically demonstrate that LFM achieves competitive generative performance compared to FM on unconditional generation of tabular and image datasets, and on conditional generation of robotic manipulation policies