Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “neural operator learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 523 records · Page 29

HydraGNN_OPF_GFM_2026 - Ensemble of predictive graph foundation models for power grid applications

This dataset supports research on graph foundation models for optimal power flow (OPF) on electric grids using HydraGNN. It contains heterogeneous graph representations of PGLib-OPF cases spanning systems from 14 to 13,659 buses, together with packed HDF5 datasets for pretraining, feasibility classification, and N-1 contingency analysis. The release includes OPF solution data, downstream fine-tuning datasets, pretrained HeteroSAGE and HeteroHEAT model checkpoints, hyperparameter-optimization summaries across multiple heterogeneous GNN architectures, and aggregated fine-tuning results for sample-efficiency studies. The dataset is designed to enable scalable training, evaluation, and transfer-learning studies for OPF surrogate modeling, including node-level AC-OPF solution prediction, graph-level prediction, feasibility classification, operating-condition generalization, and contingency-response tasks.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Fireballs in the Sky: An Augmented Reality Citizen Science Program

Fireballs in the Sky is an innovative Australian citizen science program that connects the public with the research of the Desert Fireball Network (DFN). This research aims to understand the early workings of the solar system, and Fireballs in the Sky invites people around the world to learn about this science, contributing fireball sightings via a user-friendly augmented reality mobile app. Tens of thousands of people have downloaded the app world-wide and participated in the science of meteoritics. The Fireballs in the Sky app allows users to get involved with the Desert Fireball Network research, supplementing DFN observations and providing enhanced coverage by reporting their own meteor sightings to DFN scientists. Fireballs in the Sky reports are used to track the trajectories of meteors - from their orbit in space to where they might have landed on Earth. Led by Phil Bland at Curtin University in Australia, the Desert Fireball Network (DFN) uses automated observatories across Australia to triangulate trajectories of meteorites entering the atmosphere, determine pre-entry orbits, and pinpoint their fall positions. Each observatory is an autonomous intelligent imaging system, taking 1000 by 36 megapixel all-sky images throughout the night, using neural network algorithms to recognize events. They are capable of operating for 12 months in a harsh environment, and store all imagery collected. We developed a completely automated software pipeline for data reduction, and built a supercomputer database for storage, allowing us to process our entire archive. The DFN currently stands at 50 stations distributed across the Australian continent, covering an area of 2.5 million square kilometers. Working with DFN's partners at NASA's Solar System Exploration Research Virtual Institute, the team is expanding the network beyond Australia to locations around the world. Fireballs in the Sky allows a growing public base to learn about and participate in this exciting research.

Day, Brian↗

Robustness of Deep Learning Classification to Adversarial Input on GPUs: Asynchronous Parallel Accumulation Is a Source of Vulnerability

The ability of machine learning (ML) classification models to resist small, targeted input perturbations—known as adversarial attacks—is a key measure of their safety and reliability. We show that floating-point non associativity (FPNA) coupled with asynchronous parallel programming on GPUs is sufficient to result in misclassification, without any perturbation to the input. Additionally, we show that this misclassification is particularly significant for inputs close to the decision boundary and that standard adversarial robustness results may be overestimated up to 4.6 when not considering machine-level details. We first study a linear classifier, before focusing on standard Graph Neural Network (GNN) architectures and datasets used in robustness assessments. We develop a novel black-box attack using Bayesian optimization to discover external workloads that can change the instruction scheduling which bias the output of reductions on GPUs and reliably lead to misclassification. Motivated by these results, we present a new learnable permutation (LP) gradient-based approach to learning floating-point operation orderings that lead to misclassifications. The LP approach provides a worst-case estimate in a computationally efficient manner, avoiding the need to run identical experiments tens of thousands of times over a potentially large set of possible GPU states or architectures. Finally, using instrumentation-based testing, we investigate parallel reduction ordering across different GPU architectures under external background workloads, when utilizing multi-GPU virtualization, and when applying power capping. Our results demonstrate that parallel reduction ordering varies significantly across architectures under the first two conditions, substantially increasing the search space required to fully test the effects of this parallel scheduler-based vulnerability. These results and the methods developed here can help to include machine-level considerations into adversarial robustness assessments, which can make a difference in safety and mission critical applications.

Shanmugavelu, Sanjif [Maxeler Technologies, a Groq↗

PyTorch Circuit-Aware Bit-Cell Modeling

SAND2024-08549O PyTorch Circuit-Aware Bit-Cell Modeling, a Python library, demonstrates circuit-aware training for analog machine learning accelerators. Built upon PyTorch, this software provides hardware-aware versions of common neural network layers to mitigate non-ideal behavior from novel hardware. The software provides accurate forward and backward pass estimates for hardware-mapped neural networks that are implemented in crossbars (arrays) that contain a non-linear transistor selector device. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Bennett, Christopher↗

Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units

Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Classification and Localization of Fracture-Hit Events in Low-Frequency Distributed Acoustic Sensing Strain Rate with Convolutional Neural Networks

Summary Distributed acoustic sensing (DAS) has been used in the oil and gas industry as an advanced technology for surveillance and diagnostics. Operators use DAS to monitor hydraulic fracturing activities, examine well stimulation efficacy, and estimate complex fracture system geometries. Particularly, low-frequency DAS can detect geomechanical events such as fracture hits because hydraulic fractures propagate and create strain rate variations in the rock. Analysis of DAS data today is mostly done post-job and subject to interpretation methods. However, the continuous and dense data stream generated live by DAS poses the opportunity for more efficient and accurate real-time data-driven analysis. The objective of this study is to develop a machine learning-based workflow that can identify and locate fracture-hit events in simulated strain rate responses correlated with low-frequency DAS data. In this paper, “fracture hit” refers to a hydraulic fracture originating from a stimulated well intersecting an offset well. We start with building a single fracture propagation model to produce strain rate patterns observed at a hypothetical monitoring well. This model is used to generate two sets of strain rate responses with one set containing fracture-hit events. The labeled synthetic data are then used to train a custom convolutional neural network (CNN) model for identifying the presence of fracture-hit events. The same model is trained again for locating the event with the output layer of the model replaced with linear units. We achieved near-perfect predictions for both event classification and localization. These promising results prove the feasibility of using CNN for real-time event detection from fiber-optic sensing data. Additionally, we use edge detection techniques to recognize fracture-hit event patterns in strain rate images. The fracture-hit location can be identified using recognized pixels in the image. The accuracy of edge detection-based location identification is also plausible, but edge detection is dependent on the assumption of pattern shape and image quality, hence it is less robust compared to CNN models. This comparison further supports the need for CNN applications in image-based real-time fiber-optic sensing event detection.

Engineering↗

Novel Fault Location Method for Power Systems Based on Attention Mechanism and Double Structure GRU Neural Network

Fault location is one of the most essential techniques to maintain the stable operation of power systems. A fast and accurate fault location allows operators to restore power grids faster and avoid economic losses. Conventional methods rely on expert knowledge to extract the necessary features (e.g. DWT, DFT). For large systems, more coupling effects of transmission lines require more complex feature engineering, and incomplete features can easily introduce large errors. To overcome this, a deep learning approach without manual feature extraction is introduced to the fault location model under big data application. Towards this end, in the proposed method, the attention mechanism, the Bi-GRU and a dual structure network are applied to analyze the current data from different perspectives. Complete information for the fault features is extracted for the fault location. Based on a time series model and benefit from the ability to internally acquire the information architecture of faulty line, the established model is adaptive to the power grids with very complex topologies. Simulation results indicate that the proposed double-structure model reduces the maximum error and is less affected by noise. In comparison with different structures and different models, the proposed method shows better performance in IEEE 39-bus system.

42 ENGINEERING↗

Neutrino interaction vertex reconstruction in DUNE with Pandora deep learning

The Pandora Software Development Kit and algorithm libraries perform reconstruction of neutrino interactions in liquid argon time projection chamber detectors. Pandora is the primary event reconstruction software used at the Deep Underground Neutrino Experiment, which will operate four large-scale liquid argon time projection chambers at the far detector site in South Dakota, producing high-resolution images of charged particles emerging from neutrino interactions. While these high-resolution images provide excellent opportunities for physics, the complex topologies require sophisticated pattern recognition capabilities to interpret signals from the detectors as physically meaningful objects that form the inputs to physics analyses. A critical component is the identification of the neutrino interaction vertex. Subsequent reconstruction algorithms use this location to identify the individual primary particles and ensure they each result in a separate reconstructed particle. A new vertex-finding procedure described in this article integrates a U-ResNet neural network performing hit-level classification into the multi-algorithm approach used by Pandora to identify the neutrino interaction vertex. The machine learning solution is seamlessly integrated into a chain of pattern-recognition algorithms. The technique substantially outperforms the previous BDT-based solution, with a more than 20% increase in the efficiency of sub-1 cm vertex reconstruction across all neutrino flavours.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Process prediction and detection of faults using probabilistic bidirectional recurrent neural networks on real plant data

Attaining Industry 4.0 for manufacturing operations requires advanced monitoring systems and real-time data analytics of plant data, among other topics. We propose a Probabilistic Bidirectional Recurrent Network (PBRN) for industrial process monitoring for the early detection of faults. The model is based on a Gated Recurrent Unit (GRU) neural network that allows the model to retain long-term dependencies between sensor data along a time horizon, hence learning the dynamic behavior of the process. To reduce the false-positive detection rate of the model, we compel the model to learn from a highly noisy sensor reading while outputting noise-free sensor outputs. The performance of the proposed model is compared to other data-driven statistical process monitoring schemes using real plant data from an industrial Air Separations Unit (ASU) containing noisy sensor readings. We show that the model can learn from noisy data without reducing its performance. Using two different fault cases, we demonstrate the model’s ability to carry out early fault detection with average false-positive rates of 2.9% and 4.9% for both fault cases. The missed detection rates are 0.1% and 0.2%, respectively.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Solving high-dimensional partial differential equations using deep learning

Developing algorithms for solving high-dimensional partial differential equations (PDEs) has been an exceedingly difficult task for a long time, due to the notoriously difficult problem known as the “curse of dimensionality.” This paper introduces a deep learning-based approach that can handle general high-dimensional parabolic PDEs. To this end, the PDEs are reformulated using backward stochastic differential equations and the gradient of the unknown solution is approximated by neural networks, very much in the spirit of deep reinforcement learning with the gradient acting as the policy function. Numerical results on examples including the nonlinear Black–Scholes equation, the Hamilton–Jacobi–Bellman equation, and the Allen–Cahn equation suggest that the proposed algorithm is quite effective in high dimensions, in terms of both accuracy and cost. Furthermore, this opens up possibilities in economics, finance, operational research, and physics, by considering all participating agents, assets, resources, or particles together at the same time, instead of making ad hoc assumptions on their interrelationships.

97 MATHEMATICS AND COMPUTING↗

Attack on Grid Event Cause Analysis: An Adversarial Machine Learning Approach

With the ever-increasing reliance on data for data-driven applications in power grids, such as event cause analysis, the authenticity of data streams has become crucially important. The data can be prone to adversarial stealthy attacks aiming to manipulate the data such that residual-based bad data detectors cannot detect them, and the perception of system operators or event classifiers changes about the actual event. This paper investigates the impact of adversarial attacks on convolutional neural network-based event cause analysis frameworks. We have successfully verified the ability of adversaries to maliciously misclassify events through stealthy data manipulations. The vulnerability assessment is studied with respect to the number of compromised measurements. Furthermore, a defense mechanism to robustify the performance of the event cause analysis is proposed. The effectiveness of adversarial attacks on changing the output of the framework is studied using the data generated by real-time digital simulator (RTDS) under different scenarios such as type of attacks and level of access to data.

Niazazari, Iman↗

Bayesian Segmentation (BCNN) v.1

Deep learning has been applied with great success to the segmentation of 3D X-Ray Computed Tomography (CT) scans. Establishing the credibility of these segmentations requires uncertainty quantification (UQ) to identify problem areas. Bayesian neural networks (BNNs), which use variational inference to learn the posterior distribution of the neural network weights, have been proposed to incorporate UQ into deep learning models. This software is an implementation of a novel 3D Bayesian convolutional neural network (BCNN) that provides accurate binary segmentations and uncertainty maps for 3D volumes. In particular, the uncertainty maps generated by this BCNN capture continuity and visual gradients, making them interpretable as confidence intervals for segmentation usable in numerical simulations. SAND2019-12927 M Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-NA0003525.

LaBonte, TylerMatthew↗

A Study of Physics-Informed Deep Learning for System Fluid Dynamics Closures

The sub-grid-scale (SGS) physics models, or so-called closure relations (CR), are essential in thermal-fluid modeling and simulation codes, ranging from large-eddy simulation to system-level thermo-fluid dynamic problems. While details vary, all thermo-fluid codes solve coupled mass-momentum-energy conservation partial differential equations (PDE). Traditionally, the analytical form of closure models has to be known, and then we calibrate the model using data from a set of applicable experiments. Characteristically, it takes long effort to gain insights and mechanistic understanding needed for developing closure models. Such a model is a compact form of data. The long time required for new model development limits the pace of model application while dealing with new geometries, new coolants or system conditions, particularly in newly designed systems. Alternative to mechanistic and semi-analytical models, it is possible to discover the pattern of closure relations by applying the modern machine learning methods to the body of relevant experimental data. Deep learning (DL) is a relatively new branch in the machine learning family, and a major breakthrough has been achieved since Hinton introduces a fast algorithm to efficiently train a feed-forward neural network (NN) with multilayer perceptrons. Generally, any NN with more than two layers is considered as deep learning, and it contains a large number of hyper-parameters to achieve pattern recognition or regression for complex datasets. Hornik shows that multilayer NN is compatible with universal approximation theorem. There is no constraint for a multilayer NN to capture the properties of any measurable information. This capability makes DL attractive for closure models in a system thermal-hydraulic code. As system geometry and/or operating conditions change, new physical experiments (increasingly sophisticated) are conducted to support evaluation and revision of closure models. Such a so-called data-driven modeling relaxes the structural form of system codes to directly assimilate data instead of hard-wiring the model to pre-calibrated correlations. The objective of this research is to enable cost-effective development of a DL-based closure that reflects the data. The previous work on applying machine learning techniques to fluid dynamics focused on statistical learning with basic NN from direct numerical simulations (DNS) dataset. In the present work, we show that DL-based closure models can cause the fluid calculation to diverge if multilayer NN is used. To address the issue, we investigate conditions by which the DL-based closure models work compatibly, stably, and effectively with PDE-constrained forward prediction problems. Optimally meeting the compatibility conditions constitutes a physics-informed deep learning (PIDL) strategy. We refer to this process as PDE-constrained machine learning or PDE-ML. To develop the technical basis for PDE-ML using DL-based fluid dynamics closure, numerical experiments are performed for manufactured problems of increasing complexity. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Learning the structure of wind: A data-driven nonlocal turbulence model for the atmospheric boundary layer

In this work, we develop a novel data-driven approach to modeling the atmospheric boundary layer. This approach leads to a nonlocal, anisotropic synthetic turbulence model which we refer to as the deep rapid distortion (DRD) model. Our approach relies on an operator regression problem that characterizes the best fitting candidate in a general family of nonlocal covariance kernels parameterized in part by a neural network. This family of covariance kernels is expressed in Fourier space and is obtained from approximate solutions to the Navier–Stokes equations at very high Reynolds numbers. Each member of the family incorporates important physical properties such as mass conservation and a realistic energy cascade. The DRD model can be calibrated with noisy data from field experiments. After calibration, the model can be used to generate synthetic turbulent velocity fields. To this end, we provide a new numerical method based on domain decomposition which delivers scalable, memory-efficient turbulence generation with the DRD model as well as others. We demonstrate the robustness of our approach with both filtered and noisy data coming from the 1968 Air Force Cambridge Research Laboratory Kansas experiments. Using these data, we witness exceptional accuracy with the DRD model, especially when compared to the International Electrotechnical Commission standard.

17 WIND ENERGY↗

Deep learning and radiomics: the utility of Google TensorFlow™ Inception in classifying clear cell renal cell carcinoma and oncocytoma on multiphasic CT

Purpose: Currently, all solid enhancing renal masses without microscopic fat are considered malignant until proven otherwise and there is substantial overlap in the imaging findings of benign and malignant renal masses, particularly between clear cell RCC (ccRCC) and benign oncocytoma (ONC). Radiomics has attracted increased attention for its utility in pre-operative work-up on routine clinical images. Radiomics based approaches have converted medical images into mineable data and identified prognostic imaging signatures that machine learning algorithms can use to construct predictive models by learning the decision boundaries of the underlying data distribution. The TensorFlow™ framework from Google is a state-of-the-art open-source software library that can be used for training deep learning neural networks for performing machine learning tasks. The purpose of this study was to investigate the diagnostic value and feasibility of a deep learning-based renal lesion classifier using open-source Google TensorFlow™ Inception in differentiating ccRCC from ONC on routine four-phase MDCT in patients with pathologically confirmed renal masses. Methods: With institutional review board approval for this 1996 Health Insurance Portability and Accountability Act compliant retrospective study and a waiver of informed consent, we queried our institution’s pathology, clinical, and radiology databases for histologically proven cases of ccRCC and ONC obtained between January 2000 and January 2016 scanned with a an intravenous contrast-enhanced four-phase renal mass protocol (unenhanced (UN), corticomedullary (CM), nephrographic (NP), and excretory (EX) phases). To extract features to be used for the machine learning model, the entire renal mass was contoured in the axial plane in each of the four phases, resulting in a 3D volume of interest (VOI) representative of the entire renal mass. We investigated thirteen different approaches to convert the acquired VOI data into a set of images that adequately represented each tumor which was used to train the final layer of the neural network model. Training was performed over 4000 iterations. In each iteration, 90% of the data were designated as training data and the remaining 10% served as validation data and a leave-one-out cross-validation scheme was implemented. Accuracy, sensitivity, specificity, positive (PPV) and negative predictive (NPV) values, and CIs were calculated for the classification of the thirteen processing modes. Results: We analyzed 179 consecutive patients with 179 lesions (128 ccRCC and 51 ONC). The ccRCC cohort had a mean size of 3.8 cm (range 0.8–14.6 cm) and the ONC cohort had a mean lesion size of 3.9 cm (range 1.0–13.1 cm). The highest specificity and PPV (52.9% and 80.3%, respectively) were achieved in the EX phase when we analyzed the single mid-slice of the tumor in the axial, coronal and sagittal plane, and when we increased the number of mid-slices of the tumor to three, with an accuracy of 75.4%, which also increased the sensitivity to 88.3% and the PPV to 79.6%. Using the entire tumor volume also showed that classification performance was best in the EX phase with an accuracy of 74.4%, a sensitivity of 85.8% and a PPV of 80.1%. When the entire tumor volume, plus mid-slices from all phases and all planes presented as tiled images, were submitted to the final layer of the neural network we achieved a PPV of 82.5%. Conclusions: The best classification result was obtained in the EX phase among the thirteen classification methods tested. Our proof of concept study is the first step towards understanding the utility of machine learning in the differentiation of ccRCC from ONC on routine CT images. We hope this could lead to future investigation into the development of a multivariate machine learning model which may augment our ability to accurately predict renal lesion histology on imaging.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Semantic Regression for Disentangling Beam Losses in the Fermilab Main Injector and Recycler

Fermilab’s Main Injector enclosure houses two accelerators: the Main Injector (MI) and the Recycler (RR). In periods of joint operation, when both machines contain high intensity beam, radiative beam losses from MI and RR overlap on the enclosure’s beam loss monitoring (BLM) system, making it difficult to attribute those losses to a single machine. Incorrect diagnoses result in unnecessary downtime that incurs both financial and experimental cost. In this work, we introduce a novel neural approach for automatically disentangling each machine’s contributions to those measured losses. Using a continuous adaptation of the popular UNet architecture in conjunction with a novel data augmentation scheme, our model accurately infers the machine of origin on a per-BLM basis in periods of joint and independent operation. Crucially, by extracting beam loss information at varying receptive fields, the method is capable of learning both local and global machine signatures and producing high quality inferences using only raw BLM loss measurements.

43 PARTICLE ACCELERATORS↗

Spatial Grid-Based Object Localization from A Single Passive Sensor: A Deep Learning-Integrated Approach

Ongoing efforts at NASA’s Langley Research Center have produced a single passive sensor system for detecting ground objects and pinpointing their real-world location to a desired level of precision. The Langley center serves as a test range for unmanned aerial systems (UAS) and real-time knowledge about the location of people on campus is needed to inform least-risk UAS flight operations. The proposed system provides this knowledge through a camera combined with a convolutional neural network and an algorithm that projects an imaginary grid of square cells from the ground plane onto the perspective view of the camera. The position of detected objects on the camera’s projected grid determines their location in the real-world. The imaginary grid is easily mapped to a universal coordinate system, such as longitude and latitude, to provide both relative and absolute positional information of the detected objects. This simple system is shown to be accurate and effective, with decisive advantages over alternative multi-sensor and active sensor approaches. Extensions to the system are described to allow adaptation to a variety of other use cases.

object localization↗

Energy and Emission Prediction for Mixed-Vehicle Transit Fleets Using Multi-task and Inductive Transfer Learning

Public transit agencies are focused on making their fixed-line bus systems more energy efficient by introducing electric (EV) and hybrid (HV) vehicles to their fleets. However, because of the high upfront cost of these vehicles, most agencies are tasked with managing a mixed-fleet of internal combustion vehicles (ICEVs), EVs, and HVs. In managing mixed-fleets, agencies require accurate predictions of energy use for optimizing the assignment of vehicles to transit routes, scheduling charging, and ensuring that emission standards are met. The current state-of-the-art is to develop separate neural network models to predict energy consumption for each vehicle class. Although different vehicle classes’ energy consumption depends on a varied set of covariates, we hypothesize that there are broader generalizable patterns that govern energy consumption and emissions. In this paper, we seek to extract these patterns to aid learning to address two problems faced by transit agencies. First, in the case of a transit agency which operates many ICEVs, HVs, and EVs, we use multi-task learning (MTL) to improve accuracy of forecasting energy consumption. Second, in the case where there is a significant variation in vehicles in each category, we use inductive transfer learning (ITL) to improve predictive accuracy for vehicle class models with insufficient data. As this work is to be deployed by our partner agency, we also provide an online pipeline for joining the various sensor streams for fixed-line transit energy prediction. Here, we find that our approach outperforms vehicle-specific baselines in both the MTL and ITL settings.

97 MATHEMATICS AND COMPUTING↗