We define a digital twin (DT) of a physical system governed by partial differential equations (PDEs) as a model for real-time simulations and control of the system behavior under changing conditions. We construct DTs using the Karhunen-Lo\`{e}ve Neural Network (KL-NN) surrogate model and transfer learning (TL). The surrogate model allows fast inference and differentiability with respect to control parameters for control and optimization. TL is used to retrain the model for new conditions with minimal additional data. We employ the moment equations to analyze TL and identify parameters that can be transferred to new conditions. The proposed analysis also guides the control variable selection in DT to facilitate efficient TL. For linear PDE problems, the non-transferable parameters in the KL-NN surrogate model can be exactly estimated from a single solution of the PDE corresponding to the mean values of the control variables under new target conditions. Retraining an ML model with a single solution sample is known as one-shot learning, and our analysis shows that the one-shot TL is exact for linear PDEs. For nonlinear PDE problems, transferring of any parameters introduces errors. For a nonlinear diffusion PDE model, we find that for a relatively small range of control variables, some surrogate model parameters can be transferred without introducing a significant error, some can be approximately estimated from the mean-field equation, and the rest can be found using a linear residual least square problem or an ordinary linear least square problem if a small labeled dataset for new conditions is available. The former approach results in a one-shot TL while the latter approach is an example of a few-shot TL. Both methods are approximate for the nonlinear PDEs.

Pretrained language models (LMs) can generalize to implications of facts that they are finetuned on. For example, if finetuned on ``John Doe lives in Tokyo," LMs can correctly answer ``What language do the people in John Doe's city speak?'' with ``Japanese''. However, little is known about the mechanisms that enable this generalization or how they are learned during pretraining. We introduce extractive structures as a framework for describing how components in LMs (e.g., MLPs or attention heads) coordinate to enable this generalization. The structures consist of informative components that store training facts as weight changes, and upstream and downstream extractive components that query and process the stored information to produce the correct implication. We hypothesize that extractive structures are learned during pretraining when encountering implications of previously known facts. This yields two predictions: a data ordering effect where extractive structures can be learned only if facts precede their implications, and a weight grafting effect where extractive structures can be transferred to predict counterfactual implications. We empirically demonstrate these phenomena in the OLMo-7b, Llama 3-8b, Gemma 2-9b, and Qwen 2-7b models. Of independent interest, our results also indicate that fact learning can occur at both early and late layers, which lead to different forms of generalization.

The rapid proliferation and deployment of General-Purpose AI (GPAI) models, including large language models (LLMs), present unprecedented challenges for AI supervisory entities. We hypothesize that these entities will need to navigate an emergent ecosystem of risk and incident reporting, likely to exceed their supervision capacity. To investigate this, we develop a simulation framework parameterized by features extracted from the diverse landscape of risk, incident, or hazard reporting ecosystems, including community-driven platforms, crowdsourcing initiatives, and expert assessments. We evaluate four supervision policies: non-prioritized (first-come, first-served), random selection, priority-based (addressing the highest-priority risks first), and diversity-prioritized (balancing high-priority risks with comprehensive coverage across risk types). Our results indicate that while priority-based and diversity-prioritized policies are more effective at mitigating high-impact risks, particularly those identified by experts, they may inadvertently neglect systemic issues reported by the broader community. This oversight can create feedback loops that amplify certain types of reporting while discouraging others, leading to a skewed perception of the overall risk landscape. We validate our simulation results with several real-world datasets, including one with over a million ChatGPT interactions, of which more than 150,000 conversations were identified as risky. This validation underscores the complex trade-offs inherent in AI risk supervision and highlights how the choice of risk management policies can shape the future landscape of AI risks across diverse GPAI models used in society.

Machine learning (ML) models have shown success in applications with an objective of prediction, but the algorithmic complexity of some models makes them difficult to interpret. Methods have been proposed to provide insight into these "black-box" models, but there is little research that focuses on supervised ML when the model inputs are functional data. In this work, we consider two applications from high-consequence spaces with objectives of making predictions using functional data inputs. One application aims to classify material types to identify explosive materials given hyperspectral computed tomography scans of the materials. The other application considers the forensics science task of connecting an inkjet printed document to the source printer using color signatures extracted by Raman spectroscopy. An instinctive route to consider for analyzing these data is a data driven ML model for classification, but due to the high consequence nature of the applications, we argue it is important to appropriately account for the nature of the data in the analysis to not obscure or misrepresent patterns. As such, we propose the Variable importance Explainable Elastic Shape Analysis (VEESA) pipeline for training ML models with functional data that (1) accounts for the vertical and horizontal variability in the functional data and (2) provides an explanation in the original data space of how the model uses variability in the functional data for prediction. The pipeline makes use of elastic functional principal components analysis (efPCA) to generate uncorrelated model inputs and permutation feature importance (PFI) to identify the principal components important for prediction. The variability captured by the important principal components in visualized the original data space. We ultimately discuss ideas for natural extensions of the VEESA pipeline and challenges for future research.

AAS 25-147 INVESTIGATING THE IMP ACT OF OBSERVATION SP ACE DESIGN CHOICES ON TRAINING REINFORCEMENT LEARNING SOLUTIONS FOR SP ACECRAFT PROBLEMS Nathaniel Hamilton *, Kyle Dunlap, and Kerianne L. Hobbs Recent research using Reinforcement Learning (RL) to learn autonomous control for spacecraft operations has shown great success. However, a recent study showed their performance could be improved by changing the action space, i.e. control outputs, used in the learning environment. This has opened the door for finding more improvements through further changes to the environment. The work in this paper focuses on how changes to the environment's observation space can impact the training and performance of RL agents learning the spacecraft inspection task. The studies are split into two groups. The first looks at the impact of sensors that were designed to help agents learn the task. The second looks at the impact of reference frames, reorienting the agent to see the world from a different perspective. The results show the sensors are not necessary, but most of them help agents learn more optimal behavior, and that the reference frame does not have a large impact, but is best kept consistent. INTRODUCTION Autonomous spacecraft operation is a critical capability for managing the growing number of space and increasingly complex operations.

Musculoskeletal injuries during military training significantly impact readiness, making prevention through activity monitoring crucial. While Human Activity Recognition (HAR) using wearable devices offers promising solutions, it faces challenges in processing continuous data streams and recognizing diverse activities without predefined sessions. This paper introduces an end-to-end framework for preprocessing, analyzing, and recognizing activities from wearable data in military training contexts. Using data from 135 soldiers wearing \textit{Garmin--55} smartwatches over six months with over 15 million minutes. We develop a hierarchical deep learning approach that achieves 93.8% accuracy in temporal splits and 83.8% in cross-user evaluation. Our framework addresses missing data through physiologically-informed methods, reducing unknown sleep states from 40.38% to 3.66%. We demonstrate that while longer time windows (45-60 minutes) improve basic state classification, they present trade-offs in detecting fine-grained activities. Additionally, we introduce an intuitive visualization system that enables real-time comparison of individual performance against group metrics across multiple physiological indicators. This approach to activity recognition and performance monitoring provides military trainers with actionable insights for optimizing training programs and preventing injuries.

We present a tensorization algorithm for constructing tensor train representations of functions, drawing on sketching and cross interpolation ideas. The method only requires black-box access to the target function and a small set of sample points defining the domain of interest. Thus, it is particularly well-suited for machine learning models, where the domain of interest is naturally defined by the training dataset. We show that this approach can be used to enhance the privacy and interpretability of neural network models. Specifically, we apply our decomposition to (i) obfuscate neural networks whose parameters encode patterns tied to the training data distribution, and (ii) estimate topological phases of matter that are easily accessible from the tensor train representation. Additionally, we show that this tensorization can serve as an efficient initialization method for optimizing tensor trains in general settings, and that, for model compression, our algorithm achieves a superior trade-off between memory and time complexity compared to conventional tensorization methods of neural networks.

Time series forecasting refers to a class of techniques for the prediction of events through a sequence of time, typically to inform strategic or tactical decision making. Going beyond strategic forecasting problems (e.g., those commonly-used historically in statistics and econometrics [1]), operational forecasting problems are increasingly-important. For example, at large internet retail companies, this includes demand forecasting for products at an online retailer, work force cohorts of a company in its locations, compute capacity needs per region and server type, etc.; in scientific machine learning, this includes prediction of extreme events in, e.g., climate and weather models; and so on. In particular, MQCNN [2] and MQTransformer [3] are stateof-the-art (SOTA) neural network (NN) based multivariate time series forecasting models that are used to predict future demand at the product level for hundreds of millions of products.

Understanding the motivations underlying the human inclination to automate tasks is vital to developing truly helpful robots integrated into daily life. Accordingly, we ask: are individuals more inclined to automate chores based on the time they consume or the feelings experienced while performing them? This study explores these preferences and whether they vary across different social groups (i.e., gender category and income level). Leveraging data from the BEHAVIOR-1K dataset, the American Time-Use Survey, and the American Time-Use Survey Well-Being Module, we investigate the relationship between the desire for automation, time spent on daily activities, and their associated feelings - Happiness, Meaningfulness, Sadness, Painfulness, Stressfulness, or Tiredness. Our key findings show that, despite common assumptions, time spent does not strongly relate to the desire for automation for the general population. For the feelings analyzed, only happiness and pain are key indicators. Significant differences by gender and economic level also emerged: Women prefer to automate stressful activities, whereas men prefer to automate those that make them unhappy; mid-income individuals prioritize automating less enjoyable and meaningful activities, while low and high-income show no significant correlations. We hope our research helps motivate technologies to develop robots that match the priorities of potential users, moving domestic robotics toward more socially relevant solutions. We open-source all the data, including an online tool that enables the community to replicate our analysis and explore additional trends at https://hri1260.github.io/why-automate-this.

With recent advances in Large Language Models (LLMs), Agentic AI has become phenomenal in real-world applications, moving toward multiple LLM-based agents to perceive, learn, reason, and act collaboratively. These LLM-based Multi-Agent Systems (MASs) enable groups of intelligent agents to coordinate and solve complex tasks collectively at scale, transitioning from isolated models to collaboration-centric approaches. This work provides an extensive survey of the collaborative aspect of MASs and introduces an extensible framework to guide future research. Our framework characterizes collaboration mechanisms based on key dimensions: actors (agents involved), types (e.g., cooperation, competition, or coopetition), structures (e.g., peer-to-peer, centralized, or distributed), strategies (e.g., role-based or model-based), and coordination protocols. Through a review of existing methodologies, our findings serve as a foundation for demystifying and advancing LLM-based MASs toward more intelligent and collaborative solutions for complex, real-world use cases. In addition, various applications of MASs across diverse domains, including 5G/6G networks, Industry 5.0, question answering, and social and cultural settings, are also investigated, demonstrating their wider adoption and broader impacts. Finally, we identify key lessons learned, open challenges, and potential research directions of MASs towards artificial collective intelligence.