Nathan M. Urban

Sanket Jantre, Shrijita Bhattacharya, Nathan M. Urban et al.

Deep ensembles have emerged as a powerful technique for improving predictive performance and enhancing model robustness across various applications by leveraging model diversity. However, traditional deep ensemble methods are often computationally expensive and rely on deterministic models, which may limit their flexibility. Additionally, while sparse subnetworks of dense models have shown promise in matching the performance of their dense counterparts and even enhancing robustness, existing methods for inducing sparsity typically incur training costs comparable to those of training a single dense model, as they either gradually prune the network during training or apply thresholding post-training. In light of these challenges, we propose an approach for sequential ensembling of dynamic Bayesian neural subnetworks that consistently maintains reduced model complexity throughout the training process while generating diverse ensembles in a single forward pass. Our approach involves an initial exploration phase to identify high-performing regions within the parameter space, followed by multiple exploitation phases that take advantage of the compactness of the sparse model. These exploitation phases quickly converge to different minima in the energy landscape, corresponding to high-performing subnetworks that together form a diverse and robust ensemble. We empirically demonstrate that our proposed approach outperforms traditional dense and sparse deterministic and Bayesian ensemble models in terms of prediction accuracy, uncertainty estimation, out-of-distribution detection, and adversarial robustness.

Sanket Jantre, Nathan M. Urban, Xiaoning Qian et al.

Bayesian inference for neural networks, or Bayesian deep learning, has the potential to provide well-calibrated predictions with quantified uncertainty and robustness. However, the main hurdle for Bayesian deep learning is its computational complexity due to the high dimensionality of the parameter space. In this work, we propose a novel scheme that addresses this limitation by constructing a low-dimensional subspace of the neural network parameters-referred to as an active subspace-by identifying the parameter directions that have the most significant influence on the output of the neural network. We demonstrate that the significantly reduced active subspace enables effective and scalable Bayesian inference via either Monte Carlo (MC) sampling methods, otherwise computationally intractable, or variational inference. Empirically, our approach provides reliable predictions with robust uncertainty estimates for various regression tasks.

Amir Hossein Rahmati, Mingzhou Fan, Ruida Zhou et al.

Instead of randomly acquiring training data points, Uncertainty-based Active Learning (UAL) operates by querying the label(s) of pivotal samples from an unlabeled pool selected based on the prediction uncertainty, thereby aiming at minimizing the labeling cost for model training. The efficacy of UAL critically depends on the model capacity as well as the adopted uncertainty-based acquisition function. Within the context of this study, our analytical focus is directed toward comprehending how the capacity of the machine learning model may affect UAL efficacy. Through theoretical analysis, comprehensive simulations, and empirical studies, we conclusively demonstrate that UAL can lead to worse performance in comparison with random sampling when the machine learning model class has low capacity and is unable to cover the underlying ground truth. In such situations, adopting acquisition functions that directly target estimating the prediction performance may be beneficial for improving the performance of UAL.

Alif Bin Abdul Qayyum, Xihaier Luo, Nathan M. Urban et al.

The world is moving towards clean and renewable energy sources, such as wind energy, in an attempt to reduce greenhouse gas emissions that contribute to global warming. To enhance the analysis and storage of wind data, we introduce a deep learning framework designed to simultaneously enable effective dimensionality reduction and continuous representation of multi-altitude wind data from discrete observations. The framework consists of three key components: dimensionality reduction, cross-modal prediction, and super-resolution. We aim to: (1) improve data resolution across diverse climatic conditions to recover high-resolution details; (2) reduce data dimensionality for more efficient storage of large climate datasets; and (3) enable cross-prediction between wind data measured at different heights. Comprehensive testing confirms that our approach surpasses existing methods in both super-resolution quality and compression efficiency.

Amir Hossein Rahmati, Sanket Jantre, Weifeng Zhang et al.

Low-Rank Adaptation (LoRA) offers a cost-effective solution for fine-tuning large language models (LLMs), but it often produces overconfident predictions in data-scarce few-shot settings. To address this issue, several classical statistical learning approaches have been repurposed for scalable uncertainty-aware LoRA fine-tuning. However, these approaches neglect how input characteristics affect the predictive uncertainty estimates. To address this limitation, we propose Contextual Low-Rank Adaptation (C-LoRA) as a novel uncertainty-aware and parameter efficient fine-tuning approach, by developing new lightweight LoRA modules contextualized to each input data sample to dynamically adapt uncertainty estimates. Incorporating data-driven contexts into the parameter posteriors, C-LoRA mitigates overfitting, achieves well-calibrated uncertainties, and yields robust predictions. Extensive experiments on LLaMA2-7B models demonstrate that C-LoRA consistently outperforms the state-of-the-art uncertainty-aware LoRA methods in both uncertainty quantification and model generalization. Ablation studies further confirm the critical role of our contextual modules in capturing sample-specific uncertainties. C-LoRA sets a new standard for robust, uncertainty-aware LLM fine-tuning in few-shot regimes. Although our experiments are limited to 7B models, our method is architecture-agnostic and, in principle, applies beyond this scale; studying its scaling to larger models remains an open problem. Our code is available at https://github.com/ahra99/c_lora.

Sanket Jantre, Tianle Wang, Gilchan Park et al.

Identification of protein-protein interactions (PPIs) helps derive cellular mechanistic understanding, particularly in the context of complex conditions such as neurodegenerative disorders, metabolic syndromes, and cancer. Large Language Models (LLMs) have demonstrated remarkable potential in predicting protein structures and interactions via automated mining of vast biomedical literature; yet their inherent uncertainty remains a key challenge for deriving reproducible findings, critical for biomedical applications. In this study, we present an uncertainty-aware adaptation of LLMs for PPI analysis, leveraging fine-tuned LLaMA-3 and BioMedGPT models. To enhance prediction reliability, we integrate LoRA ensembles and Bayesian LoRA models for uncertainty quantification (UQ), ensuring confidence-calibrated insights into protein behavior. Our approach achieves competitive performance in PPI identification across diverse disease contexts while addressing model uncertainty, thereby enhancing trustworthiness and reproducibility in computational biology. These findings underscore the potential of uncertainty-aware LLM adaptation for advancing precision medicine and biomedical research.

Fernando Llorente, Daniel Waxman, Sanket Jantre et al.

Gaussian processes (GPs) offer a flexible, uncertainty-aware framework for modeling complex signals, but scale cubically with data, assume static targets, and are brittle to outliers, limiting their applicability in large-scale problems with dynamic and noisy environments. Recent work introduced decentralized random Fourier feature Gaussian processes (DRFGP), an online and distributed algorithm that casts GPs in an information-filter form, enabling exact sequential inference and fully distributed computation without reliance on a fusion center. In this paper, we extend DRFGP along two key directions: first, by introducing a robust-filtering update that downweights the impact of atypical observations; and second, by incorporating a dynamic adaptation mechanism that adapts to time-varying functions. The resulting algorithm retains the recursive information-filter structure while enhancing stability and accuracy. We demonstrate its effectiveness on a large-scale Earth system application, underscoring its potential for in-situ modeling.

Sanket Jantre, Deepak Akhare, Zhiyuan Wang et al.

Partial differential equations (PDEs) underpin the modeling of many natural and engineered systems. It can be convenient to express such models as neural PDEs rather than using traditional numerical PDE solvers by replacing part or all of the PDE's governing equations with a neural network representation. Neural PDEs are often easier to differentiate, linearize, reduce, or use for uncertainty quantification than the original numerical solver. They are usually trained on solution trajectories obtained by long-horizon rollout of the PDE solver. Here we propose a more sample-efficient data-augmentation strategy for generating neural PDE training data from a computer model by space-filling sampling of local "stencil" states. This approach removes a large degree of spatiotemporal redundancy present in trajectory data and oversamples states that may be rarely visited but help the neural PDE generalize across the state space. We demonstrate that accurate neural PDE stencil operators can be learned from synthetic training data generated by the computational equivalent of 10 timesteps' worth of numerical simulation. Accuracy is further improved if we assume access to a single full-trajectory simulation from the computer model, which is typically available in practice. Across several PDE systems, we show that our data-augmented stencil data yield better trained neural stencil operators, with clear performance gains compared with naively sampled stencil data from simulation trajectories. Finally, with only 10 solver steps' worth of augmented stencil data, our approach outperforms traditional ML emulators trained on thousands of trajectories in long-horizon rollout accuracy and stability.

Amir Hossein Rahmati, Nathan M. Urban, Byung-Jun Yoon et al.

Machine Learning surrogates have been developed to accelerate solving systems dynamics of complex processes in different science and engineering applications. To faithfully capture governing systems dynamics, these methods rely on large training datasets, hence restricting their applicability in real-world problems. In this work, we propose BayPOD-AL, an active learning framework based on an uncertainty-aware Bayesian proper orthogonal decomposition (POD) approach, which aims to effectively learn reduced-order models from high-fidelity full-order models representing complex systems. Experimental results on predicting the temperature evolution over a rod demonstrate BayPOD-AL's effectiveness in suggesting the informative data and reducing computational cost related to constructing a training dataset compared to other uncertainty-guided active learning strategies. Furthermore, we demonstrate BayPOD-AL's generalizability and efficiency by evaluating its performance on a dataset of higher temporal resolution than the training dataset.

Alif Bin Abdul Qayyum, Susan D. Mertins, Amanda K. Paulson et al.

The data-driven drug design problem can be formulated as an optimization task of a potentially expensive black-box objective function over a huge high-dimensional and structured molecular space. The junction tree variational autoencoder (JTVAE) has been shown to be an efficient generative model that can be used for suggesting legitimate novel drug-like small molecules with improved properties. While the performance of the generative molecular design (GMD) scheme strongly depends on the initial training data, one can improve its sampling efficiency for suggesting better molecules with enhanced properties by optimizing the latent space. In this work, we propose how mechanistic models - such as pathway models described by differential equations - can be used for effective latent space optimization(LSO) of JTVAEs and other similar models for GMD. To demonstrate the potential of our proposed approach, we show how a pharmacodynamic model, assessing the therapeutic efficacy of a drug-like small molecule by predicting how it modulates a cancer pathway, can be incorporated for effective LSO of data-driven models for GMD.

Alif Bin Abdul Qayyum, Xihaier Luo, Nathan M. Urban et al.

World is looking for clean and renewable energy sources that do not pollute the environment, in an attempt to reduce greenhouse gas emissions that contribute to global warming. Wind energy has significant potential to not only reduce greenhouse emission, but also meet the ever increasing demand for energy. To enable the effective utilization of wind energy, addressing the following three challenges in wind data analysis is crucial. Firstly, improving data resolution in various climate conditions to ensure an ample supply of information for assessing potential energy resources. Secondly, implementing dimensionality reduction techniques for data collected from sensors/simulations to efficiently manage and store large datasets. Thirdly, extrapolating wind data from one spatial specification to another, particularly in cases where data acquisition may be impractical or costly. We propose a deep learning based approach to achieve multi-modal continuous resolution wind data prediction from discontinuous wind data, along with data dimensionality reduction.

Anthony M. DeGennaro, Nathan M. Urban

The Koopman operator is a linear, infinite-dimensional operator that governs the dynamics of system observables; Extended Dynamic Mode Decomposition (EDMD) is a data-driven method for approximating the Koopman operator using functions (features) of the system state snapshots. This paper investigates an approach to EDMD in which the features used provide random approximations to a particular kernel function. The objective of this is computational economy for large data sets: EDMD is generally ill-suited for problems with large state dimension, and its dual kernel formulation (KDMD) is well-suited for such problems only if the number of data snapshots is relatively small. We discuss two specific methods for generating features: random Fourier features, and the Nystrom method. The first method is a data-independent method for translation-invariant kernels only and involves random sampling in feature space; the second method is a data-dependent empirical method that may be used for any kernel and involves random sampling of data. We first discuss how these ideas may be applied in an EDMD context, as well as a means for adaptively adding random Fourier features. We demonstrate these methods on two example problems and conclude with an analysis of the relative benefits and drawbacks of each method.