Renewable generators must commit to day-ahead market bids despite uncertainty in both production and real-time prices. While forecasts provide valuable guidance, rare and unpredictable extreme events (so-called black swans) can cause substantial financial losses. This paper models the nomination problem as an instance of optimal transport-based distributionally robust optimization (OT-DRO), a principled framework that balances risk and performance by accounting not only for the severity of deviations but also for their likelihood. The resulting formulation yields a tractable, data-driven strategy that remains competitive under normal conditions while providing effective protection against extreme price spikes. Using four years of Finnish wind farm and market data, we demonstrate that OT-DRO consistently outperforms forecast-based nominations and significantly mitigates losses during black swan events.
A growth curve model (GCM) aims to characterize how an outcome variable evolves, develops and grows as a function of time, along with other predictors. It provides a particularly useful framework to model growth trend in longitudinal data. However, the estimation and inference of GCM with a large number of response variables faces numerous challenges, and remains underdeveloped. In this article, we study the high-dimensional multivariate-response linear GCM, and develop the corresponding estimation and inference procedures. Our proposal is far from a straightforward extension, and involves several innovative components. Specifically, we introduce a Kronecker product structure, which allows us to effectively decompose a very large covariance matrix, and to pool the correlated samples to improve the estimation accuracy. We devise a highly non-trivial multi-step estimation approach to estimate the individual covariance components separately and effectively. We also develop rigorous statistical inference procedures to test both the global effects and the individual effects, and establish the size and power properties, as well as the proper false discovery control. We demonstrate the effectiveness of the new method through both intensive simulations, and the analysis of a longitudinal neuroimaging data for Alzheimer's disease.
Nuclear power plants (NPPs) are safety–critical systems where reliable fault diagnosis and accident identification are essential for operational safety. However, in practical applications, training and deployment environments exhibit domain shift caused by both observational and evolutionary discrepancies, leading to distribution mismatch and weakening the generalization ability of conventional machine learning and domain adaptation methods. To address this challenge, this study introduces a Large Language Model-based Sequential Data framework (LLM-SD) for NPP classification tasks. The method reformulates multivariate sensor data into patch-based sequences and leverages a pre-trained large language model as a universal model of the framework. By freezing the backbone and training only lightweight embedding and classification layers, the introduced framework achieves cross-domain classification capabilities. Extensive experiments are conducted on a pressurized water reactor (PWR) simulation dataset generated using PCTRAN and the MIT Graphite Exponential Pile (MGEP) experimental dataset under multiple cross-domain scenarios. Results demonstrate that LLM-SD consistently outperforms baseline models including SVM, XGBoost, CNN, ResNet, and Transformer. The introduced framework maintains stable and robust performance across all scenarios, highlighting the effectiveness of LLM as a generalizable paradigm for classification tasks in NPPs under complex and uncertain operating conditions.
In multimodal transportation systems, shared mobility services (SMSs) are often promoted for their potential to enhance flexibility and reduce congestion. However, SMSs demand is often concentrated in high-density areas, which can limit the effectiveness and accessibility for various commuter groups across a city. This uneven integration challenges the efficiency of the transportation system, especially in terms of emissions and spatial equity. Addressing these issues requires coordination among multiple stakeholders whose objectives frequently conflict. Whereas authorities aim to ensure sustainable and equitable mobility, SMSs providers focus on revenue maximization, and travelers seek to minimize personal travel costs. This paper proposes a multi-agent deep reinforcement learning (RL) framework that captures these interactions and reconciles competing goals through dynamic pricing and incentivization strategies for SMSs and public transport. The framework integrates multimodal macroscopic simulation with two RL agents: (i) a public authority that allocates spatio-temporal public transport incentives to improve equity, emissions, and efficiency, and (ii) an SMS provider that dynamically adjusts its fares to optimize revenue. The agents interact iteratively with the simulated transportation system to learn optimal strategies in response to evolving demand, congestion, and network conditions. Numerical experiments conducted over a three-hour morning peak period, at a 20-minute temporal resolution, show that dynamic incentivization effectively reduces congestion peaks, lowers commuters’ costs by around 20% and emissions by approximately 10%, while nearly doubling public transport profit and supporting a more equitable distribution of benefits. When combined with dynamic SMS pricing, the two RL agents demonstrate the capacity to balance conflicting objectives between private providers and public authorities. The proposed approach provides a decision-support tool for guiding sustainable and equitable multimodal mobility planning.
Large language models (LLMs) are increasingly adapted to downstream tasks via reinforcement learning (RL) methods like Group Relative Policy Optimization (GRPO), which often require thousands of rollouts to learn new tasks. We argue that the interpretable nature of language often provides a much richer learning medium for LLMs, compared to policy gradients derived from sparse, scalar rewards. To test this, we introduce GEPA (Genetic-Pareto), a prompt optimizer that thoroughly incorporates natural language reflection to learn high-level rules from trial and error. Given any AI system containing one or more LLM prompts, GEPA samples trajectories (e.g., reasoning, tool calls, and tool outputs) and reflects on them in natural language to diagnose problems, propose and test prompt updates, and combine complementary lessons from the Pareto frontier of its own attempts. As a result of GEPA's design, it can often turn even just a few rollouts into a large quality gain. Across four tasks, GEPA outperforms GRPO by 6% on average and by up to 20%, while using up to 35x fewer rollouts. GEPA also outperforms the leading prompt optimizer, MIPROv2, by over 10% (e.g., +10% accuracy on AIME-2025).
