Artur Dubrawski

Chirag Nagpal, Willa Potosnak, Artur Dubrawski

Applications of machine learning in healthcare often require working with time-to-event prediction tasks including prognostication of an adverse event, re-hospitalization or death. Such outcomes are typically subject to censoring due to loss of follow up. Standard machine learning methods cannot be applied in a straightforward manner to datasets with censored outcomes. In this paper, we present auton-survival, an open-source repository of tools to streamline working with censored time-to-event or survival data. auton-survival includes tools for survival regression, adjustment in the presence of domain shift, counterfactual estimation, phenotyping for risk stratification, evaluation, as well as estimation of treatment effects. Through real world case studies employing a large subset of the SEER oncology incidence data, we demonstrate the ability of auton-survival to rapidly support data scientists in answering complex health and epidemiological questions.

Malgorzata Gwiazda, Yifu Cai, Mononito Goswami et al. · cmu

Large Language Models (LLMs) have shown promising performance in time series modeling tasks, but do they truly understand time series data? While multiple benchmarks have been proposed to answer this fundamental question, most are manually curated and focus on narrow domains or specific skill sets. To address this limitation, we propose scalable methods for creating comprehensive time series reasoning benchmarks that combine the flexibility of templates with the creativity of LLM agents. We first develop TimeSeriesExam, a multiple-choice benchmark using synthetic time series to evaluate LLMs across five core reasoning categories: pattern recognitionnoise understandingsimilarity analysisanomaly detection, and causality. Then, with TimeSeriesExamAgent, we scale our approach by automatically generating benchmarks from real-world datasets spanning healthcare, finance and weather domains. Through multi-dimensional quality evaluation, we demonstrate that our automatically generated benchmarks achieve diversity comparable to manually curated alternatives. However, our experiments reveal that LLM performance remains limited in both abstract time series reasoning and domain-specific applications, highlighting ongoing challenges in enabling effective time series understanding in these models. TimeSeriesExamAgent is available at https://github.com/magwiazda/TimeSeriesExamAgent.

Kin G. Olivares, Azul Garza, David Luo et al.

Large collections of time series data are commonly organized into structures with different levels of aggregation; examples include product and geographical groupings. It is often important to ensure that the forecasts are coherent so that the predicted values at disaggregate levels add up to the aggregate forecast. The growing interest of the Machine Learning community in hierarchical forecasting systems indicates that we are in a propitious moment to ensure that scientific endeavors are grounded on sound baselines. For this reason, we put forward the HierarchicalForecast library, which contains preprocessed publicly available datasets, evaluation metrics, and a compiled set of statistical baseline models. Our Python-based reference framework aims to bridge the gap between statistical and econometric modeling, and Machine Learning forecasting research. Code and documentation are available in https://github.com/Nixtla/hierarchicalforecast.

Mononito Goswami, Vedant Sanil, Arjun Choudhry et al. · cmu

Machine learning (ML) models are only as good as the data they are trained on. But recent studies have found datasets widely used to train and evaluate ML models, e.g. ImageNet, to have pervasive labeling errors. Erroneous labels on the train set hurt ML models' ability to generalize, and they impact evaluation and model selection using the test set. Consequently, learning in the presence of labeling errors is an active area of research, yet this field lacks a comprehensive benchmark to evaluate these methods. Most of these methods are evaluated on a few computer vision datasets with significant variance in the experimental protocols. With such a large pool of methods and inconsistent evaluation, it is also unclear how ML practitioners can choose the right models to assess label quality in their data. To this end, we propose a benchmarking environment AQuA to rigorously evaluate methods that enable machine learning in the presence of label noise. We also introduce a design space to delineate concrete design choices of label error detection models. We hope that our proposed design space and benchmark enable practitioners to choose the right tools to improve their label quality and that our benchmark enables objective and rigorous evaluation of machine learning tools facing mislabeled data.

Arnab Dey, Mononito Goswami, Joo Heung Yoon et al. · cmu

A significant proportion of clinical physiologic monitoring alarms are false. This often leads to alarm fatigue in clinical personnel, inevitably compromising patient safety. To combat this issue, researchers have attempted to build Machine Learning (ML) models capable of accurately adjudicating Vital Sign (VS) alerts raised at the bedside of hemodynamically monitored patients as real or artifact. Previous studies have utilized supervised ML techniques that require substantial amounts of hand-labeled data. However, manually harvesting such data can be costly, time-consuming, and mundane, and is a key factor limiting the widespread adoption of ML in healthcare (HC). Instead, we explore the use of multiple, individually imperfect heuristics to automatically assign probabilistic labels to unlabeled training data using weak supervision. Our weakly supervised models perform competitively with traditional supervised techniques and require less involvement from domain experts, demonstrating their use as efficient and practical alternatives to supervised learning in HC applications of ML.

Chufan Gao, Mononito Goswami, Jieshi Chen et al. · cmu

Healthcare providers usually record detailed notes of the clinical care delivered to each patient for clinical, research, and billing purposes. Due to the unstructured nature of these narratives, providers employ dedicated staff to assign diagnostic codes to patients' diagnoses using the International Classification of Diseases (ICD) coding system. This manual process is not only time-consuming but also costly and error-prone. Prior work demonstrated potential utility of Machine Learning (ML) methodology in automating this process, but it has relied on large quantities of manually labeled data to train the models. Additionally, diagnostic coding systems evolve with time, which makes traditional supervised learning strategies unable to generalize beyond local applications. In this work, we introduce a general weakly-supervised text classification framework that learns from class-label descriptions only, without the need to use any human-labeled documents. It leverages the linguistic domain knowledge stored within pre-trained language models and the data programming framework to assign code labels to individual texts. We demonstrate the efficacy and flexibility of our method by comparing it to state-of-the-art weak text classifiers across four real-world text classification datasets, in addition to assigning ICD codes to medical notes in the publicly available MIMIC-III database.

Michał Wiliński, Mononito Goswami, Willa Potosnak et al. · cmu

Time series foundation models (TSFMs) promise to be powerful tools for a wide range of applications. However, their internal representations and learned concepts are still not well understood. In this study, we investigate the structure and redundancy of representations across various TSFMs, examining the self-similarity of model layers within and across different model sizes. This analysis reveals block-like redundancy in the representations, which can be utilized for informed pruning to improve inference speed and efficiency. Additionally, we explore the concepts learned by these models - such as periodicity and trends - and how these can be manipulated through latent space steering to influence model behavior. Our experiments show that steering interventions can introduce new features, e.g., adding periodicity or trends to signals that initially lacked them. These findings underscore the value of representational analysis for optimizing models and demonstrate how conceptual steering offers new possibilities for more controlled and efficient time series analysis with TSFMs.

Cecilia Morales, Jason Yao, Tejas Rane et al.

Robot-guided catheter insertion has the potential to deliver urgent medical care in situations where medical personnel are unavailable. However, this technique requires accurate and reliable segmentation of anatomical landmarks in the body. For the ultrasound imaging modality, obtaining large amounts of training data for a segmentation model is time-consuming and expensive. This paper introduces RESUS (RESlicing of UltraSound Images), a weak supervision data augmentation technique for ultrasound images based on slicing reconstructed 3D volumes from tracked 2D images. This technique allows us to generate views which cannot be easily obtained in vivo due to physical constraints of ultrasound imaging, and use these augmented ultrasound images to train a semantic segmentation model. We demonstrate that RESUS achieves statistically significant improvement over training with non-augmented images and highlight qualitative improvements through vessel reconstruction.

Willa Potosnak, Cristian Challu, Mononito Goswami et al. · cmu

Recently, time series foundation models have shown promising zero-shot forecasting performance on time series from a wide range of domains. However, it remains unclear whether their success stems from a true understanding of temporal dynamics or simply from memorizing the training data. While implicit reasoning in language models has been studied, similar evaluations for time series models have been largely unexplored. This work takes an initial step toward assessing the reasoning abilities of deep time series forecasting models. We find that certain linear, MLP-based, and patch-based Transformer models generalize effectively in systematically orchestrated out-of-distribution scenarios, suggesting underexplored reasoning capabilities beyond simple pattern memorization.

Nina Żukowska, Mononito Goswami, Michał Wiliński et al. · cmu

Time series foundation models have shown impressive performance on a variety of tasks, across a wide range of domains, even in zero-shot settings. However, most of these models are designed to handle short univariate time series as an input. This limits their practical use, especially in domains such as healthcare with copious amounts of long and multivariate data with strong temporal and intra-variate dependencies. Our study bridges this gap by cataloging and systematically comparing various context expansion techniques from both language and time series domains, and introducing a novel compressive memory mechanism to allow encoder-only TSFMs to effectively model intra-variate dependencies. We demonstrate the benefits of our approach by imbuing MOMENT, a recent family of multi-task time series foundation models, with the multivariate context.

Benedikt Boecking, Nicholas Roberts, Willie Neiswanger et al.

Many promising applications of supervised machine learning face hurdles in the acquisition of labeled data in sufficient quantity and quality, creating an expensive bottleneck. To overcome such limitations, techniques that do not depend on ground truth labels have been studied, including weak supervision and generative modeling. While these techniques would seem to be usable in concert, improving one another, how to build an interface between them is not well-understood. In this work, we propose a model fusing programmatic weak supervision and generative adversarial networks and provide theoretical justification motivating this fusion. The proposed approach captures discrete latent variables in the data alongside the weak supervision derived label estimate. Alignment of the two allows for better modeling of sample-dependent accuracies of the weak supervision sources, improving the estimate of unobserved labels. It is the first approach to enable data augmentation through weakly supervised synthetic images and pseudolabels. Additionally, its learned latent variables can be inspected qualitatively. The model outperforms baseline weak supervision label models on a number of multiclass image classification datasets, improves the quality of generated images, and further improves end-model performance through data augmentation with synthetic samples.

Willa Potosnak, Nina Å»ukowska, Michał Wiliński et al. · cmu

Multivariate forecasting with Transformers faces a core scalability challenge: modeling cross-channel dependencies via attention compounds attention's quadratic sequence complexity with quadratic channel scaling, making full cross-channel attention impractical for high-dimensional time series. We propose Multivariate Infini Compressive Attention (MICA), an architectural design to extend channel-independent Transformers to channel-dependent forecasting. By adapting efficient attention techniques from the sequence dimension to the channel dimension, MICA adds a cross-channel attention mechanism to channel-independent backbones that scales linearly with channel count and context length. We evaluate channel-independent Transformer architectures with and without MICA across multiple forecasting benchmarks. MICA reduces forecast error over its channel-independent counterparts by 5.4% on average and up to 25.4% on individual datasets, highlighting the importance of explicit cross-channel modeling. Moreover, models with MICA rank first among deep multivariate Transformer and MLP baselines. MICA models also scale more efficiently with respect to both channel count and context length than Transformer baselines that compute attention across both the temporal and channel dimensions, establishing compressive attention as a practical solution for scalable multivariate forecasting.

Brian Kunzer, Mario Berges, Artur Dubrawski

The concept of a digital twin has exploded in popularity over the past decade, yet confusion around its plurality of definitions, its novelty as a new technology, and its practical applicability still exists, all despite numerous reviews, surveys, and press releases. The history of the term digital twin is explored, as well as its initial context in the fields of product life cycle management, asset maintenance, and equipment fleet management, operations, and planning. A definition for a minimally viable framework to utilize a digital twin is also provided based on seven essential elements. A brief tour through DT applications and industries where DT methods are employed is also outlined. The application of a digital twin framework is highlighted in the field of predictive maintenance, and its extensions utilizing machine learning and physics based modeling. Employing the combination of machine learning and physics based modeling to form hybrid digital twin frameworks, may synergistically alleviate the shortcomings of each method when used in isolation. Key challenges of implementing digital twin models in practice are additionally discussed. As digital twin technology experiences rapid growth and as it matures, its great promise to substantially enhance tools and solutions for intelligent upkeep of complex equipment, are expected to materialize.

Mononito Goswami, Konrad Szafer, Arjun Choudhry et al. · cmu

We introduce MOMENT, a family of open-source foundation models for general-purpose time series analysis. Pre-training large models on time series data is challenging due to (1) the absence of a large and cohesive public time series repository, and (2) diverse time series characteristics which make multi-dataset training onerous. Additionally, (3) experimental benchmarks to evaluate these models, especially in scenarios with limited resources, time, and supervision, are still in their nascent stages. To address these challenges, we compile a large and diverse collection of public time series, called the Time series Pile, and systematically tackle time series-specific challenges to unlock large-scale multi-dataset pre-training. Finally, we build on recent work to design a benchmark to evaluate time series foundation models on diverse tasks and datasets in limited supervision settings. Experiments on this benchmark demonstrate the effectiveness of our pre-trained models with minimal data and task-specific fine-tuning. Finally, we present several interesting empirical observations about large pre-trained time series models. Pre-trained models (AutonLab/MOMENT-1-large) and Time Series Pile (AutonLab/Timeseries-PILE) are available on Huggingface.

Benedikt Boecking, Vincent Jeanselme, Artur Dubrawski

Clustering under pairwise constraints is an important knowledge discovery tool that enables the learning of appropriate kernels or distance metrics to improve clustering performance. These pairwise constraints, which come in the form of must-link and cannot-link pairs, arise naturally in many applications and are intuitive for users to provide. However, the common practice of relaxing discrete constraints to a continuous domain to ease optimization when learning kernels or metrics can harm generalization, as information which only encodes linkage is transformed to informing distances. We introduce a new constrained clustering algorithm that jointly clusters data and learns a kernel in accordance with the available pairwise constraints. To generalize well, our method is designed to maximize constraint satisfaction without relaxing pairwise constraints to a continuous domain where they inform distances. We show that the proposed method outperforms existing approaches on a large number of diverse publicly available datasets, and we discuss how our method can scale to handling large data.

Chirag Nagpal, Vedant Sanil, Artur Dubrawski

Studies involving both randomized experiments as well as observational data typically involve time-to-event outcomes such as time-to-failure, death or onset of an adverse condition. Such outcomes are typically subject to censoring due to loss of follow-up and established statistical practice involves comparing treatment efficacy in terms of hazard ratios between the treated and control groups. In this paper we propose a statistical approach to recovering sparse phenogroups (or subtypes) that demonstrate differential treatment effects as compared to the study population. Our approach involves modelling the data as a mixture while enforcing parameter shrinkage through structured sparsity regularization. We propose a novel inference procedure for the proposed model and demonstrate its efficacy in recovering sparse phenotypes across large landmark real world clinical studies in cardiovascular health.

Yifu Cai, Arjun Choudhry, Mononito Goswami et al. · cmu

Large Language Models (LLMs) have recently demonstrated a remarkable ability to model time series data. These capabilities can be partly explained if LLMs understand basic time series concepts. However, our knowledge of what these models understand about time series data remains relatively limited. To address this gap, we introduce TimeSeriesExam, a configurable and scalable multiple-choice question exam designed to assess LLMs across five core time series understanding categories: pattern recognition, noise understanding, similarity analysis, anomaly detection, and causality analysis. TimeSeriesExam comprises of over 700 questions, procedurally generated using 104 carefully curated templates and iteratively refined to balance difficulty and their ability to discriminate good from bad models. We test 7 state-of-the-art LLMs on the TimeSeriesExam and provide the first comprehensive evaluation of their time series understanding abilities. Our results suggest that closed-source models such as GPT-4 and Gemini understand simple time series concepts significantly better than their open-source counterparts, while all models struggle with complex concepts such as causality analysis. We believe that the ability to programatically generate questions is fundamental to assessing and improving LLM's ability to understand and reason about time series data.

Maria De-Arteaga, Alexandra Chouldechova, Artur Dubrawski

Effective human-AI collaboration requires a system design that provides humans with meaningful ways to make sense of and critically evaluate algorithmic recommendations. In this paper, we propose a way to augment human-AI collaboration by building on a common organizational practice: identifying experts who are likely to provide complementary opinions. When machine learning algorithms are trained to predict human-generated assessments, experts' rich multitude of perspectives is frequently lost in monolithic algorithmic recommendations. The proposed approach aims to leverage productive disagreement by (1) identifying whether some experts are likely to disagree with an algorithmic assessment and, if so, (2) recommend an expert to request a second opinion from.

Cecilia G. Morales, Dhruv Srikanth, Jack H. Good et al.

In trauma and critical care settings, rapid and precise intravascular access is key to patients' survival. Our research aims at ensuring this access, even when skilled medical personnel are not readily available. Vessel bifurcations are anatomical landmarks that can guide the safe placement of catheters or needles during medical procedures. Although ultrasound is advantageous in navigating anatomical landmarks in emergency scenarios due to its portability and safety, to our knowledge no existing algorithm can autonomously extract vessel bifurcations using ultrasound images. This is primarily due to the limited availability of ground truth data, in particular, data from live subjects, needed for training and validating reliable models. Researchers often resort to using data from anatomical phantoms or simulations. We introduce BIFURC, Bifurcation Identification for Ultrasound-driven Robot Cannulation, a novel algorithm that identifies vessel bifurcations and provides optimal needle insertion sites for an autonomous robotic cannulation system. BIFURC integrates expert knowledge with deep learning techniques to efficiently detect vessel bifurcations within the femoral region and can be trained on a limited amount of in-vivo data. We evaluated our algorithm using a medical phantom as well as real-world experiments involving live pigs. In all cases, BIFURC consistently identified bifurcation points and needle insertion locations in alignment with those identified by expert clinicians.

Brad Shook, Abby Turner, Jieshi Chen et al.

Time series foundation models (TSFMs) pretrained on data from multiple domains have shown strong performance on diverse modeling tasks. Various efforts have been made to develop foundation models specific to electroencephalography (EEG) data, which records brain electrical activity as time series. However, no comparative analysis of EEG-specific foundation models (EEGFMs) versus general TSFMs has been performed on EEG-specific tasks. We introduce a novel Spatial-Temporal Adapter with Multi-Head Pooling (STAMP), which leverages univariate embeddings produced by a general TSFM, implicitly models spatial-temporal characteristics of EEG data, and achieves performance comparable to state-of-the-art EEGFMs. A comprehensive analysis is performed on 8 benchmark datasets of clinical tasks using EEG for classification, along with ablation studies. Our proposed adapter is lightweight in trainable parameters and flexible in the inputs it can accommodate, supporting easy modeling of EEG data using TSFMs.

Yifu Cai, Xinyu Li, Mononito Goswami et al. · cmu

We introduce TimeSeriesGym, a scalable benchmarking framework for evaluating Artificial Intelligence (AI) agents on time series machine learning engineering challenges. Existing benchmarks lack scalability, focus narrowly on model building in well-defined settings, and evaluate only a limited set of research artifacts (e.g., CSV submission files). To make AI agent benchmarking more relevant to the practice of machine learning engineering, our framework scales along two critical dimensions. First, recognizing that effective ML engineering requires a range of diverse skills, TimeSeriesGym incorporates challenges from diverse sources spanning multiple domains and tasks. We design challenges to evaluate both isolated capabilities (including data handling, understanding research repositories, and code translation) and their combinations, and rather than addressing each challenge independently, we develop tools that support designing multiple challenges at scale. Second, we implement evaluation mechanisms for multiple research artifacts, including submission files, code, and models, using both precise numeric measures and more flexible LLM-based evaluation approaches. This dual strategy balances objective assessment with contextual judgment. Although our initial focus is on time series applications, our framework can be readily extended to other data modalities, broadly enhancing the comprehensiveness and practical utility of agentic AI evaluation. We open-source our benchmarking framework to facilitate future research on the ML engineering capabilities of AI agents.

Willa Potosnak, Cristian Challu, Mononito Goswami et al. · cmu

Large pre-trained time series foundation models (TSFMs) have demonstrated promising zero-shot performance across a wide range of domains. However, a question remains: Do TSFMs succeed by memorizing patterns in training data, or do they possess the ability to reason about such patterns? While reasoning is a topic of great interest in the study of Large Language Models (LLMs), it is undefined and largely unexplored in the context of TSFMs. In this work, inspired by language modeling literature, we formally define compositional reasoning in forecasting and distinguish it from in-distribution generalization. We evaluate the reasoning and generalization capabilities of 16 popular deep learning forecasting models on multiple synthetic and real-world datasets. Additionally, through controlled studies, we systematically examine which design choices in 7 popular open-source TSFMs contribute to improved reasoning capabilities. Our study yields key insights into the impact of TSFM architecture design on compositional reasoning and generalization. We find that patch-based Transformers have the best reasoning performance, closely followed by residualized MLP-based architectures, which are 97\% less computationally complex in terms of FLOPs and 86\% smaller in terms of the number of trainable parameters. Interestingly, in some zero-shot out-of-distribution scenarios, these models can outperform moving average and exponential smoothing statistical baselines trained on in-distribution data. Only a few design choices, such as the tokenization method, had a significant (negative) impact on Transformer model performance.

Szymon Rusiecki, Cecilia Morales, Pia Störy et al.

Autonomous robots deployed in mass casualty incidents (MCI) face the challenge of making critical decisions based on incomplete and noisy perceptual data. We present an autonomous robotic system for casualty assessment that fuses outputs from multiple vision-based algorithms, estimating signs of severe hemorrhage, visible trauma, or physical alertness, into a coherent triage assessment. At the core of our system is a Bayesian network, constructed from expert-defined rules, which enables probabilistic reasoning about a casualty's condition even with missing or conflicting sensory inputs. The system, evaluated during the DARPA Triage Challenge (DTC) in realistic MCI scenarios involving 11 and 9 casualties, demonstrated a nearly three-fold improvement in physiological assessment accuracy (from 15\% to 42\% and 19\% to 46\%) compared to a vision-only baseline. More importantly, overall triage accuracy increased from 14\% to 53\%, while the diagnostic coverage of the system expanded from 31\% to 95\% of cases. These results demonstrate that integrating expert-guided probabilistic reasoning with advanced vision-based sensing can significantly enhance the reliability and decision-making capabilities of autonomous systems in critical real-world applications.

Chufan Gao, Nicholas Gisolfi, Artur Dubrawski

Signal quality assessment (SQA) is required for monitoring the reliability of data acquisition systems, especially in AI-driven Predictive Maintenance (PMx) application contexts. SQA is vital for addressing "silent failures" of data acquisition hardware and software, which when unnoticed, misinform the users of data, creating the risk for incorrect decisions with unintended or even catastrophic consequences. We have developed an open-source software implementation of signal quality indices (SQIs) for the analysis of time-series data. We codify a range of SQIs, demonstrate them using established benchmark data, and show that they can be effective for signal quality assessment. We also study alternative approaches to denoising time-series data in an attempt to improve the quality of the already degraded signal, and evaluate them empirically on relevant real-world data. To our knowledge, our software toolkit is the first to provide an open source implementation of a broad range of signal quality assessment and improvement techniques validated on publicly available benchmark data for ease of reproducibility. The generality of our framework can be easily extended to assessing reliability of arbitrary time-series measurements in complex systems, especially when morphological patterns of the waveform shapes and signal periodicity are of key interest in downstream analyses.

Willa Potosnak, Cristian Challu, Kin G. Olivares et al.

Forecasting healthcare time series data is vital for early detection of adverse outcomes and patient monitoring. However, it can be challenging in practice due to variable medication administration and unique pharmacokinetic (PK) properties of each patient. To address these challenges, we propose a novel hybrid global-local architecture and a PK encoder that informs deep learning models of patient-specific treatment effects. We showcase the efficacy of our approach in achieving significant accuracy gains in a blood glucose forecasting task using both realistically simulated and real-world data. Our PK encoder surpasses baselines by up to 16.4% on simulated data and 4.9% on real-world data for individual patients during critical events of severely high and low glucose levels. Furthermore, our proposed hybrid global-local architecture outperforms patient-specific PK models by 15.8%, on average.

Kin G. Olivares, David Luo, Cristian Challu et al.

Large collections of time series data are often organized into hierarchies with different levels of aggregation; examples include product and geographical groupings. Probabilistic coherent forecasting is tasked to produce forecasts consistent across levels of aggregation. In this study, we propose to augment neural forecasting architectures with a coherent multivariate mixture output. We optimize the networks with a composite likelihood objective, allowing us to capture time series' relationships while maintaining high computational efficiency. Our approach demonstrates 13.2% average accuracy improvements on most datasets compared to state-of-the-art baselines. We conduct ablation studies of the framework components and provide theoretical foundations for them. To assist related work, the code is available at this https://github.com/Nixtla/neuralforecast.

Cristian Challu, Kin G. Olivares, Boris N. Oreshkin et al.

Recent progress in neural forecasting accelerated improvements in the performance of large-scale forecasting systems. Yet, long-horizon forecasting remains a very difficult task. Two common challenges afflicting the task are the volatility of the predictions and their computational complexity. We introduce N-HiTS, a model which addresses both challenges by incorporating novel hierarchical interpolation and multi-rate data sampling techniques. These techniques enable the proposed method to assemble its predictions sequentially, emphasizing components with different frequencies and scales while decomposing the input signal and synthesizing the forecast. We prove that the hierarchical interpolation technique can efficiently approximate arbitrarily long horizons in the presence of smoothness. Additionally, we conduct extensive large-scale dataset experiments from the long-horizon forecasting literature, demonstrating the advantages of our method over the state-of-the-art methods, where N-HiTS provides an average accuracy improvement of almost 20% over the latest Transformer architectures while reducing the computation time by an order of magnitude (50 times). Our code is available at bit.ly/3VA5DoT

Kin G. Olivares, Cristian Challu, Grzegorz Marcjasz et al.

We extend the neural basis expansion analysis (NBEATS) to incorporate exogenous factors. The resulting method, called NBEATSx, improves on a well performing deep learning model, extending its capabilities by including exogenous variables and allowing it to integrate multiple sources of useful information. To showcase the utility of the NBEATSx model, we conduct a comprehensive study of its application to electricity price forecasting (EPF) tasks across a broad range of years and markets. We observe state-of-the-art performance, significantly improving the forecast accuracy by nearly 20% over the original NBEATS model, and by up to 5% over other well established statistical and machine learning methods specialized for these tasks. Additionally, the proposed neural network has an interpretable configuration that can structurally decompose time series, visualizing the relative impact of trend and seasonal components and revealing the modeled processes' interactions with exogenous factors. To assist related work we made the code available in https://github.com/cchallu/nbeatsx.

Ignacy Stępka, Nicholas Gisolfi, Artur Dubrawski

Recent advancements in machine learning have accelerated its widespread adoption across various real-world applications. However, in safety-critical domains, the deployment of machine learning models is riddled with challenges due to their complexity, lack of interpretability, and absence of formal guarantees regarding their behavior. In this paper, we introduce a verification framework tailored for Bayesian networks, designed to address these drawbacks. Our framework comprises two key components: (1) a two-step compilation and encoding scheme that translates Bayesian networks into Boolean logic literals, and (2) formal verification queries that leverage these literals to verify various properties encoded as constraints. Specifically, we introduce two verification queries: if-then rules (ITR) and feature monotonicity (FMO). We benchmark the efficiency of our verification scheme and demonstrate its practical utility in real-world scenarios.

Szymon Rusiecki, Cecilia G. Morales, Kimberly Elenberg et al.

Mass Casualty Incidents can overwhelm emergency medical systems and resulting delays or errors in the assessment of casualties can lead to preventable deaths. We present a decision support framework that fuses outputs from multiple computer vision models, estimating signs of severe hemorrhage, respiratory distress, physical alertness, or visible trauma, into a Bayesian network constructed entirely from expert-defined rules. Unlike traditional data-driven models, our approach does not require training data, supports inference with incomplete information, and is robust to noisy or uncertain observations. We report performance for two missions involving 11 and 9 casualties, respectively, where our Bayesian network model substantially outperformed vision-only baselines during evaluation of our system in the DARPA Triage Challenge (DTC) field scenarios. The accuracy of physiological assessment improved from 15% to 42% in the first scenario and from 19% to 46% in the second, representing nearly threefold increase in performance. More importantly, overall triage accuracy increased from 14% to 53% in all patients, while the diagnostic coverage of the system expanded from 31% to 95% of the cases requiring assessment. These results demonstrate that expert-knowledge-guided probabilistic reasoning can significantly enhance automated triage systems, offering a promising approach to supporting emergency responders in MCIs. This approach enabled Team Chiron to achieve 4th place out of 11 teams during the 1st physical round of the DTC.

Rohini Banerjee, Cecilia G. Morales, Artur Dubrawski

Efficient intravascular access in trauma and critical care significantly impacts patient outcomes. However, the availability of skilled medical personnel in austere environments is often limited. Autonomous robotic ultrasound systems can aid in needle insertion for medication delivery and support non-experts in such tasks. Despite advances in autonomous needle insertion, inaccuracies in vessel segmentation predictions pose risks. Understanding the uncertainty of predictive models in ultrasound imaging is crucial for assessing their reliability. We introduce MSU-Net, a novel multistage approach for training an ensemble of U-Nets to yield accurate ultrasound image segmentation maps. We demonstrate substantial improvements, 18.1% over a single Monte Carlo U-Net, enhancing uncertainty evaluations, model transparency, and trustworthiness. By highlighting areas of model certainty, MSU-Net can guide safe needle insertions, empowering non-experts to accomplish such tasks.

Jinmyeong Choi, Brad Shook, Artur Dubrawski

Time series foundation models (TSFMs) are widely used as generic feature extractors, yet the notion of non-stationarity in their embedding spaces remains poorly understood. Recent work often conflates non-stationarity with distribution shift, blurring distinctions fundamental to classical time-series analysis and long-standing methodologies such as statistical process control (SPC). In SPC, non-stationarity signals a process leaving a stable regime - via shifts in mean, variance, or emerging trends - and detecting such departures is central to quality monitoring and change-point analysis. Motivated by this diagnostic tradition, we study how different forms of distributional non-stationarity - mean shifts, variance changes, and linear trends - become linearly accessible in TSFM embedding spaces under controlled conditions. We further examine temporal non-stationarity arising from persistence, which reflects violations of weak stationarity due to long-memory or near-unit-root behavior rather than explicit distributional shifts. By sweeping shift strength and probing multiple TSFMs, we find that embedding-space detectability of non-stationarity degrades smoothly and that different models exhibit distinct, model-specific failure modes.

Cecilia G. Morales, Fanurs Chi En Teh, Kai Li et al.

Remote photoplethysmography (rPPG) enables contactless vital sign monitoring using standard RGB cameras. However, existing methods rely on fixed parameters optimized for particular lighting conditions and camera setups, limiting adaptability to diverse deployment environments. This paper introduces the Projection-based Robust Signal Mixing (PRISM) algorithm, a training-free method that jointly optimizes photometric detrending and color mixing through online parameter adaptation based on signal quality assessment. PRISM achieves state-of-the-art performance among unsupervised methods, with MAE of 0.77 bpm on PURE and 0.66 bpm on UBFC-rPPG, and accuracy of 97.3\% and 97.5\% respectively at a 5 bpm threshold. Statistical analysis confirms PRISM performs equivalently to leading supervised methods ($p > 0.2$), while maintaining real-time CPU performance without training. This validates that adaptive time series optimization significantly improves rPPG across diverse conditions.

Ignacy Stępka, Nicholas Gisolfi, Kacper Trębacz et al.

We consider the problem of persistent client dropout in asynchronous Decentralized Federated Learning (DFL). Asynchronicity and decentralization obfuscate information about model updates among federation peers, making recovery from a client dropout difficult. Access to the number of learning epochs, data distributions, and all the information necessary to precisely reconstruct the missing neighbor's loss functions is limited. We show that obvious mitigations do not adequately address the problem and introduce adaptive strategies based on client reconstruction. We show that these strategies can effectively recover some performance loss caused by dropout. Our work focuses on asynchronous DFL with local regularization and differs substantially from that in the existing literature. We evaluate the proposed methods on tabular and image datasets, involve three DFL algorithms, and three data heterogeneity scenarios (iid, non-iid, class-focused non-iid). Our experiments show that the proposed adaptive strategies can be effective in maintaining robustness of federated learning, even if they do not reconstruct the missing client's data precisely. We also discuss the limitations and identify future avenues for tackling the problem of client dropout.

Chang Liu, Jieshi Chen, Lee H. Harrison et al. · cmu

When selecting data to build machine learning models in practical applications, factors such as availability, acquisition cost, and discriminatory power are crucial considerations. Different data modalities often capture unique aspects of the underlying phenomenon, making their utilities complementary. On the other hand, some sources of data host structural information that is key to their value. Hence, the utility of one data type can sometimes be enhanced by matching the structure of another. We propose Multimodal Structure Preservation Learning (MSPL) as a novel method of learning data representations that leverages the clustering structure provided by one data modality to enhance the utility of data from another modality. We demonstrate the effectiveness of MSPL in uncovering latent structures in synthetic time series data and recovering clusters from whole genome sequencing and antimicrobial resistance data using mass spectrometry data in support of epidemiology applications. The results show that MSPL can imbue the learned features with external structures and help reap the beneficial synergies occurring across disparate data modalities.

Ifigeneia Apostolopoulou, Benjamin Eysenbach, Frank Nielsen et al.

In supervised learning, understanding an input's proximity to the training data can help a model decide whether it has sufficient evidence for reaching a reliable prediction. While powerful probabilistic models such as Gaussian Processes naturally have this property, deep neural networks often lack it. In this paper, we introduce Distance Aware Bottleneck (DAB), i.e., a new method for enriching deep neural networks with this property. Building on prior information bottleneck approaches, our method learns a codebook that stores a compressed representation of all inputs seen during training. The distance of a new example from this codebook can serve as an uncertainty estimate for the example. The resulting model is simple to train and provides deterministic uncertainty estimates by a single forward pass. Finally, our method achieves better out-of-distribution (OOD) detection and misclassification prediction than prior methods, including expensive ensemble methods, deep kernel Gaussian Processes, and approaches based on the standard information bottleneck.

Chirag Nagpal, Mononito Goswami, Keith Dufendach et al.

Estimation of treatment efficacy of real-world clinical interventions involves working with continuous outcomes such as time-to-death, re-hospitalization, or a composite event that may be subject to censoring. Counterfactual reasoning in such scenarios requires decoupling the effects of confounding physiological characteristics that affect baseline survival rates from the effects of the interventions being assessed. In this paper, we present a latent variable approach to model heterogeneous treatment effects by proposing that an individual can belong to one of latent clusters with distinct response characteristics. We show that this latent structure can mediate the base survival rates and helps determine the effects of an intervention. We demonstrate the ability of our approach to discover actionable phenotypes of individuals based on their treatment response on multiple large randomized clinical trials originally conducted to assess appropriate treatments to reduce cardiovascular risk.

Mononito Goswami, Benedikt Boecking, Artur Dubrawski

Analysing electrocardiograms (ECGs) is an inexpensive and non-invasive, yet powerful way to diagnose heart disease. ECG studies using Machine Learning to automatically detect abnormal heartbeats so far depend on large, manually annotated datasets. While collecting vast amounts of unlabeled data can be straightforward, the point-by-point annotation of abnormal heartbeats is tedious and expensive. We explore the use of multiple weak supervision sources to learn diagnostic models of abnormal heartbeats via human designed heuristics, without using ground truth labels on individual data points. Our work is among the first to define weak supervision sources directly on time series data. Results show that with as few as six intuitive time series heuristics, we are able to infer high quality probabilistic label estimates for over 100,000 heartbeats with little human effort, and use the estimated labels to train competitive classifiers evaluated on held out test data.

Piotr S. Maciąg, Robert Bembenik, Artur Dubrawski

In this article, we introduce a novel type of spatio-temporal sequential patterns called Constricted Spatio-Temporal Sequential (CSTS) patterns and thoroughly analyze their properties. We demonstrate that the set of CSTS patterns is a concise representation of all spatio-temporal sequential patterns that can be discovered in a given dataset. To measure significance of the discovered CSTS patterns we adapt the participation index measure. We also provide CSTS-Miner: an algorithm that discovers all participation index strong CSTS patterns in event data. We experimentally evaluate the proposed algorithms using two crime-related datasets: Pittsburgh Police Incident Blotter Dataset and Boston Crime Incident Reports Dataset. In the experiments, the CSTS-Miner algorithm is compared with the other four state-of-the-art algorithms: STS-Miner, CSTPM, STBFM and CST-SPMiner. As the results of experiments suggest, the proposed algorithm discovers much fewer patterns than the other selected algorithms. Finally, we provide the examples of interesting crime-related patterns discovered by the proposed CSTS-Miner algorithm.

Cecilia G. Morales, Nicholas Gisolfi, Robert Edman et al.

We recommend using a model-centric, Boolean Satisfiability (SAT) formalism to obtain useful explanations of trained model behavior, different and complementary to what can be gleaned from LIME and SHAP, popular data-centric explanation tools in Artificial Intelligence (AI). We compare and contrast these methods, and show that data-centric methods may yield brittle explanations of limited practical utility. The model-centric framework, however, can offer actionable insights into risks of using AI models in practice. For critical applications of AI, split-second decision making is best informed by robust explanations that are invariant to properties of data, the capability offered by model-centric frameworks.

Salva Rühling Cachay, Benedikt Boecking, Artur Dubrawski

Aggregating multiple sources of weak supervision (WS) can ease the data-labeling bottleneck prevalent in many machine learning applications, by replacing the tedious manual collection of ground truth labels. Current state of the art approaches that do not use any labeled training data, however, require two separate modeling steps: Learning a probabilistic latent variable model based on the WS sources -- making assumptions that rarely hold in practice -- followed by downstream model training. Importantly, the first step of modeling does not consider the performance of the downstream model. To address these caveats we propose an end-to-end approach for directly learning the downstream model by maximizing its agreement with probabilistic labels generated by reparameterizing previous probabilistic posteriors with a neural network. Our results show improved performance over prior work in terms of end model performance on downstream test sets, as well as in terms of improved robustness to dependencies among weak supervision sources.

Salva Rühling Cachay, Benedikt Boecking, Artur Dubrawski

Data programming (DP) has proven to be an attractive alternative to costly hand-labeling of data. In DP, users encode domain knowledge into \emph{labeling functions} (LF), heuristics that label a subset of the data noisily and may have complex dependencies. A label model is then fit to the LFs to produce an estimate of the unknown class label. The effects of label model misspecification on test set performance of a downstream classifier are understudied. This presents a serious awareness gap to practitioners, in particular since the dependency structure among LFs is frequently ignored in field applications of DP. We analyse modeling errors due to structure over-specification. We derive novel theoretical bounds on the modeling error and empirically show that this error can be substantial, even when modeling a seemingly sensible structure.

Cristian Challu, Kin G. Olivares, Gus Welter et al.

Neural forecasting has shown significant improvements in the accuracy of large-scale systems, yet predicting extremely long horizons remains a challenging task. Two common problems are the volatility of the predictions and their computational complexity; we addressed them by incorporating smoothness regularization and mixed data sampling techniques to a well-performing multi-layer perceptron based architecture (NBEATS). We validate our proposed method, DMIDAS, on high-frequency healthcare and electricity price data with long forecasting horizons (~1000 timestamps) where we improve the prediction accuracy by 5% over state-of-the-art models, reducing the number of parameters of NBEATS by nearly 70%.

Maria De-Arteaga, Vincent Jeanselme, Artur Dubrawski et al.

Machine learning (ML) is increasingly being used to support high-stakes decisions. However, there is frequently a construct gap: a gap between the construct of interest to the decision-making task and what is captured in proxies used as labels to train ML models. As a result, ML models may fail to capture important dimensions of decision criteria, hampering their utility for decision support. Thus, an essential step in the design of ML systems for decision support is selecting a target label among available proxies. In this work, we explore the use of historical expert decisions as a rich -- yet also imperfect -- source of information that can be combined with observed outcomes to narrow the construct gap. We argue that managers and system designers may be interested in learning from experts in instances where they exhibit consistency with each other, while learning from observed outcomes otherwise. We develop a methodology to enable this goal using information that is commonly available in organizational information systems. This involves two core steps. First, we propose an influence function-based methodology to estimate expert consistency indirectly when each case in the data is assessed by a single expert. Second, we introduce a label amalgamation approach that allows ML models to simultaneously learn from expert decisions and observed outcomes. Our empirical evaluation, using simulations in a clinical setting and real-world data from the child welfare domain, indicates that the proposed approach successfully narrows the construct gap, yielding better predictive performance than learning from either observed outcomes or expert decisions alone.

Benedikt Boecking, Willie Neiswanger, Eric Xing et al.

Obtaining large annotated datasets is critical for training successful machine learning models and it is often a bottleneck in practice. Weak supervision offers a promising alternative for producing labeled datasets without ground truth annotations by generating probabilistic labels using multiple noisy heuristics. This process can scale to large datasets and has demonstrated state of the art performance in diverse domains such as healthcare and e-commerce. One practical issue with learning from user-generated heuristics is that their creation requires creativity, foresight, and domain expertise from those who hand-craft them, a process which can be tedious and subjective. We develop the first framework for interactive weak supervision in which a method proposes heuristics and learns from user feedback given on each proposed heuristic. Our experiments demonstrate that only a small number of feedback iterations are needed to train models that achieve highly competitive test set performance without access to ground truth training labels. We conduct user studies, which show that users are able to effectively provide feedback on heuristics and that test set results track the performance of simulated oracles.

Ifigeneia Apostolopoulou, Elan Rosenfeld, Artur Dubrawski

The Variational Autoencoder (VAE) is a powerful framework for learning probabilistic latent variable generative models. However, typical assumptions on the approximate posterior distribution of the encoder and/or the prior, seriously restrict its capacity for inference and generative modeling. Variational inference based on neural autoregressive models respects the conditional dependencies of the exact posterior, but this flexibility comes at a cost: such models are expensive to train in high-dimensional regimes and can be slow to produce samples. In this work, we introduce an orthogonal solution, which we call self-reflective inference. By redesigning the hierarchical structure of existing VAE architectures, self-reflection ensures that the stochastic flow preserves the factorization of the exact posterior, sequentially updating the latent codes in a recurrent manner consistent with the generative model. We empirically demonstrate the clear advantages of matching the variational posterior to the exact posterior - on binarized MNIST, self-reflective inference achieves state-of-the art performance without resorting to complex, computationally expensive components such as autoregressive layers. Moreover, we design a variational normalizing flow that employs the proposed architecture, yielding predictive benefits compared to its purely generative counterpart. Our proposed modification is quite general and complements the existing literature; self-reflective inference can naturally leverage advances in distribution estimation and generative modeling to improve the capacity of each layer in the hierarchy.

Yichong Xu, Ruosong Wang, Lin F. Yang et al.

Preference-based Reinforcement Learning (PbRL) replaces reward values in traditional reinforcement learning by preferences to better elicit human opinion on the target objective, especially when numerical reward values are hard to design or interpret. Despite promising results in applications, the theoretical understanding of PbRL is still in its infancy. In this paper, we present the first finite-time analysis for general PbRL problems. We first show that a unique optimal policy may not exist if preferences over trajectories are deterministic for PbRL. If preferences are stochastic, and the preference probability relates to the hidden reward values, we present algorithms for PbRL, both with and without a simulator, that are able to identify the best policy up to accuracy $\varepsilon$ with high probability. Our method explores the state space by navigating to under-explored states, and solves PbRL using a combination of dueling bandits and policy search. Experiments show the efficacy of our method when it is applied to real-world problems.

Kyle Miller, Artur Dubrawski

This paper reviews current literature in the field of predictive maintenance from the system point of view. We differentiate the existing capabilities of condition estimation and failure risk forecasting as currently applied to simple components, from the capabilities needed to solve the same tasks for complex assets. System-level analysis faces more complex latent degradation states, it has to comprehensively account for active maintenance programs at each component level and consider coupling between different maintenance actions, while reflecting increased monetary and safety costs for system failures. As a result, methods that are effective for forecasting risk and informing maintenance decisions regarding individual components do not readily scale to provide reliable sub-system or system level insights. A novel holistic modeling approach is needed to incorporate available structural and physical knowledge and naturally handle the complexities of actively fielded and maintained assets.

Chirag Nagpal, Xinyu Rachel Li, Artur Dubrawski

We describe a new approach to estimating relative risks in time-to-event prediction problems with censored data in a fully parametric manner. Our approach does not require making strong assumptions of constant proportional hazard of the underlying survival distribution, as required by the Cox-proportional hazard model. By jointly learning deep nonlinear representations of the input covariates, we demonstrate the benefits of our approach when used to estimate survival risks through extensive experimentation on multiple real world datasets with different levels of censoring. We further demonstrate advantages of our model in the competing risks scenario. To the best of our knowledge, this is the first work involving fully parametric estimation of survival times with competing risks in the presence of censoring.

Benedikt Boecking, Artur Dubrawski

The scalability of the labeling process and the attainable quality of labels have become limiting factors for many applications of machine learning. The programmatic creation of labeled datasets via the synthesis of noisy heuristics provides a promising avenue to address this problem. We propose to improve modeling of latent class variables in the programmatic creation of labeled datasets by incorporating pairwise feedback into the process. We discuss the ease with which such pairwise feedback can be obtained or generated in many application domains. Our experiments show that even a small number of sources of pairwise feedback can substantially improve the quality of the posterior estimate of the latent class variable.