Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “time series prediction”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Hydrologic Model Data for the East Fork Poplar Creek Watershed Simulated with the Advanced Terrestrial Simulator (ATS): Streamflow and Network Expansion–Contraction Dynamics

This dataset supports hydrologic modeling and stream network expansion–contraction analysis for the East Fork Poplar Creek (EFPC) Watershed in Tennessee. It includes a Jupyter notebook for model setup, model configuration files, simulation outputs, and derived products used to evaluate model performance and investigate stream dynamics under varying hydrologic conditions. The dataset was generated using the Watershed Workflow Python package and the Advanced Terrestrial Simulator (ATS), enabling integrated surface–subsurface hydrologic simulations using a stream-aligned mesh. Outputs include high-resolution time series of streamflow, active network length, water table depth, and related hydrologic variables. Also included are spatially explicit stream persistency indices and classifications of reaches as perennial or non-perennial. These data facilitate reproducibility and support further research on stream intermittency and variability in network extent.The model data archive is organized in following directories:1) model_setup_inputsContains the Watershed Workflow Jupyter notebooks (accessed through any open source code editor), selected input datasets, and resulting ATS input files, including XML files (access through any open source code editor), computational mesh (.exo files can be viewed using Paraview), and meteorological forcing files (.h5 files can be accessed through h5py python package and HDFView open source software). 2) model_outputsIncludes ATS simulation outputs relevant to this study. Time series of spatially integrated or averaged variables (e.g., streamflow, water table depth) are provided as CSV files. Select spatial fields (e.g., ponded depth and water table depth) are saved as pickled Python objects to reduce file size, and can be accessed through pickle package in Python. Key geometry objects from Watershed Workflow—such as the surface mesh and river tree—are also included to support analysis of streamflow persistency and expansion–contraction dynamics. These files can also be accessed through Watershed Workflow Python package.3) model_evaluationProvides observed streamflow time series and field survey-based flow regime classifications used to evaluate model performance. Jupyter notebooks for processing ATS outputs and comparing model predictions with observations to build confidence in the model prior to scientific analysis are also included.4) Q_L_relationshipsContains workflows for generating time series of discharge, active network length, and related hydrologic variables used in the stream network expansion–contraction analysis. Includes routines for delineating baseflow-dominated periods. For each catchment, notebooks and processed data (as pickled DataFrames accessed through Pandas Python package) are provided. 5) figure_scriptsProvides the Jupyter notebooks used to generate the figures presented in the paper.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Deterministic incident-wave elevation prediction in intermediate water depth

Potential performance gains from optimal (non-causal) impedance-matching control of wave energy devices in irregular ocean waves are dependent on deterministic wave elevation prediction techniques that work well in practical applications. Although a number of devices are designed for operation in intermediate water depths, little work has been reported on deterministic wave prediction in such depths. Here, we investigated a deterministic wave-prediction technique based on an approximate propagation model that leads to an analytical formulation, which may be convenient to implement in practice. To improve accuracy, an approach to combine predictions based on multiple up-wave measurement points is evaluated. The overall method is tested using experimental time-series measurements recorded in the U.S. Navy MASK basin in Carderock, MD, USA. For comparison, an alternative prediction approach based on Fourier coefficients is also tested with the same data. Comparison of prediction approaches with direct measurements suggest room for improvement. Possible sources of error including tank reflections are estimated, and potential mitigation approaches are discussed.

16 TIDAL AND WAVE POWER↗
NASA NTRSDate not supplied

Communication Demand in the National Airspace - A Federated Learning Approach

Within the national airspace system (NAS), efficient use of spectrum remains a challenge; as UAS and UAM missions evolve, the amount of mission-critical aircraft communications are expected to significantly grow. To accommodate the increased demand, NASA Glenn Research Center is investigating artificial intelligence approaches that could dynamically allocate spectrum; however, these solutions are driven by communication and aviation data items, many of which are not directly available. One such cornerstone data item is communication demand, parameterizing the needs within a sector in terms that may directly inform spectrum allocation, such as channel access duration, bandwidth, and modulation type. This paper considers the complexity of predicting communication demand as a function of NAS behaviors. Unlike prior prediction work in communications, this research must consider how the NAS may be impacted by external factors - such as convective weather and closures - rather than estimating demand from time-series forecasting alone. Much of this research considers a federated learning design to predict communication events in terms of the type of event occurring (sector coordination, conflict resolution, etc.). To do so, an investigation of products from Sherlock Data Warehouse is conducted, identifying the trends, sufficiency, and correlations of each product to identified events. Additionally, a preliminary discussion for inferring associations between these event types and their communication parameters (duration, bandwidth, modulation) is presented. By utilizing federated learning, imbalances in the types of events and data present throughout the NAS can inform local models without impairing global training. Furthermore, the two-stage approach proposed allows for robust and speculative communication modelling, as communication techniques continue to evolve. As a result, this model enables a generalized approach to understanding NAS communications which is able to inform long-term changes to aviation spectrum management.

Nathan Schimpf↗
NASA NTRSDate not supplied

A Federated Learning Approach to Predicting Communication Demand in the National Airspace

Within the national airspace system (NAS), efficient use of spectrum remains a challenge; as UAS and UAM missions evolve, the amount of mission-critical aircraft communications are expected to significantly grow. To accommodate the increased demand, NASA Glenn Research Center is investigating artificial intelligence approaches that could dynamically allocate spectrum; however, these solutions are driven by communication and aviation data items, many of which are not directly available. One such cornerstone data item is communication demand, parameterizing the needs within a sector in terms that may directly inform spectrum allocation, such as channel access duration, bandwidth, and modulation type. This paper considers the complexity of predicting communication demand as a function of NAS behaviors. Unlike prior prediction work in communications, this research must consider how the NAS may be impacted by external factors - such as convective weather and closures - rather than estimating demand from time-series forecasting alone. Much of this research considers a federated learning design to predict communication events in terms of the type of event occurring (sector coordination, conflict resolution, etc). To do so, an investigation of products from Sherlock Data Warehouse is conducted, identifying the trends, sufficiency, and correlations of each product to identified events. Additionally, a preliminary discussion for inferring associations between these event types and their communication parameters (duration, bandwidth, modulation) is presented. By utilizing federated learning, imbalances in the types of events and data present throughout the NAS can inform local models without impairing global training. Furthermore, the two-stage approach proposed allows for robust and speculative communication modelling, as communication techniques continue to evolve. As a result, this model enables a generalized approach to understanding NAS communications which is able to inform long-term changes to aviation spectrum management.

Nathan Schimpf↗
DOE OSTI2020

Automated fault detection of residential air-conditioning systems using thermostat drive cycles

Residential air conditioning equipment comprises a significant portion of the total energy consumption of a home. Unfortunately, air-conditioning systems can be susceptible to faulty operation either from installation errors or faults that accrue over the equipment’s lifetime. This paper presents a novel automated fault detection algorithm for residential air-conditioning systems that can alert the homeowner of the presence of these faults. The proposed algorithm utilizes only the home’s thermostat and outside air temperature to perform automated fault detection over the course of the equipment’s lifetime, including immediately after installation. The algorithm uses an extended Kalman filter approach to identify a three-resistor, two-capacitor (3R2C) electrical equivalent thermodynamic model. The identified 3R2C model is used to predict cooling times during a testing period comprising of a series of thermostat drive-cycle experiments. We tested the algorithm on an EnergyPlus™ model of a typical residential building in Orlando, Florida. Duct faults, indoor airflow faults, and refrigerant undercharge faults were introduced into the building model one at a time. The algorithm was able to accurately determine duct-leak faults, 40% airflow faults, 40% undercharge faults, and no-fault cases with an accuracy of 70%, 77%, 82%, and 87%, respectively.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2020

Autotrophic and mixotrophic metabolism of an anammox bacterium revealed by in vivo 13 C and 2 H metabolic network mapping

Anaerobic ammonium-oxidizing (anammox) bacteria mediate a key step in the biogeochemical nitrogen cycle and have been applied worldwide for the energy-efficient removal of nitrogen from wastewater. However, outside their core energy metabolism, little is known about the metabolic networks driving anammox bacterial anabolism and use of different carbon and energy substrates beyond genome-based predictions. Here, we experimentally resolved the central carbon metabolism of the anammox bacterium Candidatus ‘Kuenenia stuttgartiensis’ using time-series 13 C and 2 H isotope tracing, metabolomics, and isotopically nonstationary metabolic flux analysis. Our findings confirm predicted metabolic pathways used for CO 2 fixation, central metabolism, and amino acid biosynthesis in K. stuttgartiensis , and reveal several instances where genomic predictions are not supported by in vivo metabolic fluxes. This includes the use of the oxidative branch of an incomplete tricarboxylic acid cycle for alpha-ketoglutarate biosynthesis, despite the genome not having an annotated citrate synthase. We also demonstrate that K. stuttgartiensis is able to directly assimilate extracellular formate via the Wood–Ljungdahl pathway instead of oxidizing it completely to CO 2 followed by reassimilation. In contrast, our data suggest that K. stuttgartiensis is not capable of using acetate as a carbon or energy source in situ and that acetate oxidation occurred via the metabolic activity of a low-abundance microorganism in the bioreactor’s side population. Together, these findings provide a foundation for understanding the carbon metabolism of anammox bacteria at a systems-level and will inform future studies aimed at elucidating factors governing their function and niche differentiation in natural and engineered ecosystems.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1998

Large-Scale Dynamics of the Magnetospheric Boundary: Comparisons between Global MHD Simulation Results and ISTP Observations

Understanding the large-scale dynamics of the magnetospheric boundary is an important step towards achieving the ISTP mission's broad objective of assessing the global transport of plasma and energy through the geospace environment. Our approach is based on three-dimensional global magnetohydrodynamic (MHD) simulations of the solar wind-magnetosphere- ionosphere system, and consists of using interplanetary magnetic field (IMF) and plasma parameters measured by solar wind monitors upstream of the bow shock as input to the simulations for predicting the large-scale dynamics of the magnetospheric boundary. The validity of these predictions is tested by comparing local data streams with time series measured by downstream spacecraft crossing the magnetospheric boundary. In this paper, we review results from several case studies which confirm that our MHD model reproduces very well the large-scale motion of the magnetospheric boundary. The first case illustrates the complexity of the magnetic field topology that can occur at the dayside magnetospheric boundary for periods of northward IMF with strong Bx and By components. The second comparison reviewed combines dynamic and topological aspects in an investigation of the evolution of the distant tail at 200 R(sub E) from the Earth.

Berchem, J.↗
NASA NTRS2011

Assessment of an Automated Touchdown Detection Algorithm for the Orion Crew Module

Orion Crew Module (CM) touchdown detection is critical to activating the post-landing sequence that safe?s the Reaction Control Jets (RCS), ensures that the vehicle remains upright, and establishes communication with recovery forces. In order to accommodate safe landing of an unmanned vehicle or incapacitated crew, an onboard automated detection system is required. An Orion-specific touchdown detection algorithm was developed and evaluated to differentiate landing events from in-flight events. The proposed method will be used to initiate post-landing cutting of the parachute riser lines, to prevent CM rollover, and to terminate RCS jet firing prior to submersion. The RCS jets continue to fire until touchdown to maintain proper CM orientation with respect to the flight path and to limit impact loads, but have potentially hazardous consequences if submerged while firing. The time available after impact to cut risers and initiate the CM Up-righting System (CMUS) is measured in minutes, whereas the time from touchdown to RCS jet submersion is a function of descent velocity, sea state conditions, and is often less than one second. Evaluation of the detection algorithms was performed for in-flight events (e.g. descent under chutes) using hi-fidelity rigid body analyses in the Decelerator Systems Simulation (DSS), whereas water impacts were simulated using a rigid finite element model of the Orion CM in LS-DYNA. Two touchdown detection algorithms were evaluated with various thresholds: Acceleration magnitude spike detection, and Accumulated velocity changed (over a given time window) spike detection. Data for both detection methods is acquired from an onboard Inertial Measurement Unit (IMU) sensor. The detection algorithms were tested with analytically generated in-flight and landing IMU data simulations. The acceleration spike detection proved to be faster while maintaining desired safety margin. Time to RCS jet submersion was predicted analytically across a series of simulated Orion landing conditions. This paper details the touchdown detection method chosen and the analysis used to support the decision.

Gay, Robert S.↗
NASA NTRS2015

ACCEPT: Introduction of the Adverse Condition and Critical Event Prediction Toolbox

The prediction of anomalies or adverse events is a challenging task, and there are a variety of methods which can be used to address the problem. In this paper, we introduce a generic framework developed in MATLAB (sup registered mark) called ACCEPT (Adverse Condition and Critical Event Prediction Toolbox). ACCEPT is an architectural framework designed to compare and contrast the performance of a variety of machine learning and early warning algorithms, and tests the capability of these algorithms to robustly predict the onset of adverse events in any time-series data generating systems or processes.

machine learning↗
NASA NTRS2019

FUN3D and USM3D Analysis of the Propulsion Aerodynamic Workshop 2018 S-Duct Test Case

This work presents the results of Fun3D and USM3D analyses that were performed for the 4th AIAA Propulsion Aerodynamics Workshop (PAW). The PAW workshop is separated into three sections that focus on internal duct flows, nozzle flows and a special topic. This paper focuses on the internal duct flow section of PAW04 while an accompanying paper discusses the analyses performed for the nozzle portion. For the internal duct flow section, the PAW04 participants were provided with the two configurations consisting of an S-duct with and without aerodynamic interface plane (AIP) rake legs modeled. The participants were asked to perform a grid refinement study as well as a turbulence model study for the configuration with the rake legs. The analyses discussed here were performed on custom grids developed under the guidelines of the workshop. Additionally, the paper discusses the development and use of flow controllers for matching the desired flow characteristics. The results show that both solvers do well for predicting internal flow characteristics of the S-duct based on direct comparison with the experimental data. However, the CFD-to-CFD comparison proved to be more challenging due to the localized occurrence of supersonic flow near the rake legs when using the mass flow controller. A turbulence model study was performed to compare the two-equation SST model to the SA-QCR model. The results show that although the turbulence model does affect the solution, it makes a minimal impact on pressure recovery and inlet distortion intensity for this case. Suggestions for future workshops include gridding guidelines similar to those employed for the Drag Prediction Workshop series for the grid refinement study and a time accuracy study.

Carter, Melissa B.↗
NASA NTRS2022

Bryce Canyon Water Resources: Monitoring Vegetation Health and Water Availability in Bryce Canyon National Park for Drought Stress Mitigation Planning

Bryce Canyon National Park is home to groundwater-dependent ecosystems (GDEs) that are threatened by a multidecadal drought and increased groundwater extraction due to a spike in tourism. These ecosystems contain unique species that are only found in areas where near-surface groundwater is present, such as aspen groves and fens. These species contribute to the high biodiversity found in Bryce Canyon, which boosts an ecosystem’s productivity and the services it provides to the park. Unfortunately, many of these GDEs are too small to identify with traditional Earth observation platforms and are difficult to physically reach for monitoring purposes. This project partnered with the National Park Service to identify springs and seeps as a proxy for GDEs within Bryce Canyon from 2013–2022. Furthermore, this project tested the feasibility of various methods to detect and monitor springs and seeps and therefore facilitate the partner’s efforts to conserve these ecologically valuable GDEs in Bryce Canyon. The team mapped groundwater discharge with high resolution National Agriculture Imagery Program (NAIP) and assessed park vegetation trends with Landsat 8 Operational Land Imager (OLI) and PlanetScope imagery. In-situ precipitation data and the Western Land Data Assimilation System (WLDAS) were used to produce time series of climatic variables. Seeps and spring locations were predicted using random forest classification and maximum entropy machine learning models.

Groundwater dependent ecosystems↗
DOE OSTI2022

Multimodal sensor fusion framework for residential building occupancy detection

For several years now, smart building energy systems have been a research area of intensive activity. In light of the increasing need for sustainable buildings and energy systems, this trend motivates an increasing need for a solution to reduce carbon dioxide emissions and improve energy efficiency. This work proposes a high-performing and transferable occupancy detection framework that combines sensor data from different data modalities, including time series environmental data (temperature, humidity, and illuminance), image data, and acoustic energy data using ensemble method. To draw out the best prediction performance in each modality, the proposed framework was developed, including various models that were designed to learn the occupancy patterns reflected in the physical data streams. To tackle the time series environmental data, we designed two variants of an occupancy detection spatiotemporal pattern network (Occ-STPN) that performs both feature level and decision level fusion, respectively. We also propose a new metric; the fading memory mean square error (FMMSE), that provides a fair evaluation and penalization of delayed occupancy predictions. Multiple open-sourced datasets, including the Electricity Consumption and Occupancy and the University of California, Irvine's (UCI) building occupancy detection dataset, along with our own real data collected from six different houses, were used to validate the algorithms' performance. The experimental results presented herein break down the performance for each sensing modality, and a detailed analysis of the performance is also discussed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Sub-daily virus sampling at the Bermuda Atlantic Time Series reveals diel and depth-structured population dynamics without community-level shifts

Ocean microbes contribute to biogeochemical cycles and ecosystem function, but they do so under top-down pressure imposed by viruses. While viruses are increasingly understood spatially and beginning to be incorporated into predictive modeling, high-frequency ocean virus dynamics remain understudied due to methodological challenges. Here we sampled stratified Bermuda Atlantic Time Series (BATS) waters for 112 hours at sub-daily 4- (surface) or 12- (deep chlorophyll maximum) hour intervals, purified viral particles from these samples, sequenced their metagenomes, and used the resulting data to characterize high-frequency virus community dynamics. Aggregated community diversity metrics changed with depth, but were not statistically significant temporally at a fixed location. However, finer-scale population-level analyses revealed both depth and temporal change, including physicochemical depth-driven differences and, in surface waters, thousands of viral populations that exhibited statistically significant diel rhythms. Statistical analyses revealed three main archetypes of temporal dynamics that themselves differed in abundance patterns, host predictions, viral taxonomy, and gene functions. Among these, highlights include viruses resembling an archetype with a night peaking pattern in activity that include an over-representation of viruses that putatively infect Prochlorococcus, a phototrophic cyanobacteria. Together, these efforts provide baseline community- and population-scale short-time-frame observations relevant to future climate state modeling.

Carrillo, Alfonso [The Ohio State University, Colu↗
DOE OSTI2026

Safe Deep Reinforcement Learning for Active Distribution System Model Predictive Control with EVs and DERs

The temporal and spatial mismatch between PV generation and electric vehicle (EV) charging and discharging may cause voltage violations in active distribution networks. Despite the widespread use of deep reinforcement learning (DRL) in power system optimization and control, it lacks guarantees on constraint satisfaction during both training and deployment. This paper proposes a Lagrangian-based safe DRL approach for model predictive control (MPC) of active distribution systems with large-scale integration of PVs, EVs, and energy storage systems (ESSs). A Transformer-LSTM time-series model is proposed to forecast EV charging demand, which is then formulated as a constraint to ensure charging requirements are met. Using this prediction, a Lagrangian-based safe soft actor-critic (SAC) framework is developed for real-time control in a three-phase unbalanced distribution system, enforcing voltage safety constraints while optimizing the cumulative net reward. By integrating the forecasting model with multi-period constraints, the proposed framework jointly coordinates PV systems, EV charging and discharging, and ESS scheduling within the MPC horizon. Numerical experiments on a modified IEEE 123-bus system with real-world data show that, under a high PV penetration scenario, the proposed method increases the net reward by 30.74% and reduces average voltage violations from 0.0011 p.u. to 0.0002 p.u. compared with standard SAC. Compared with the optimal power flow (OPF) approach, it achieves similar voltage security while yielding lower line losses. It also maintains real-time control capability, reducing operation latency to 53.21 ms per 15-minute control interval. The proposed method remains effective under varying PV/EV penetrations and load conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2020

Verification of Prognostic Algorithms to Predict Remaining Flying Time for Electric Unmanned Vehicles

This paper addresses the problem of building trust in the online prediction of a eUAV’s remaining available flying time powered by lithium-ion polymer batteries. A series of ground tests are described that make use of an electric unmanned aerial vehicle (eUAV) to verify the performance of remaining flying time predictions. The algorithm verification procedure described is implemented on a fully functional vehicle that is restrained to a platform for repeated run-to-functional-failure (charge depletion) experiments. The vehicle under test is commanded to follow a predefined propeller RPM profile in order to create battery demand profiles similar to those expected during flight. The eUAV is repeatedly operated until the charge stored in powertrain batteries falls below a specified limit threshold. The time at which the limit threshold on battery charge is crossed is then used to measure the accuracy of the remaining flying time prediction. In our earlier work battery aging was not included. In this work we take into account aging of the batteries where the parameters were updated to make predictions. Accuracy requirements are considered for an alarm that warns operators when remaining flying time is estimated to fall below the specified limit threshold.

Edward F. Hogge↗
DOE OSTI2022

The L-CAPE Project at FNAL

The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigating the cause of an outage and labeling it after the fact. However, as one of the most upstream systems at the FNAL accelerator complex, the Linac’s foreknowledge of an impending downtime as well as its duration could prompt downstream systems to go into standby, potentially leading to energy savings. The goals of the Linac Condition Anomaly Prediction of Emergence (L-CAPE) project that started in late 2020 are (1) to apply data-analytic methods to improve the information that is available to operators in the control room, and (2) to use machine learning to automate the labeling of outage types as they occur and discover patterns in the data that could lead to the prediction of outages. We present an overview of the challenges in dealing with time-series data from 2000+ devices, our approach to developing an ML-based automated outage labeling system, and the status of augmenting operations by identifying the most likely devices predicting an outage.

43 PARTICLE ACCELERATORS↗
DOE OSTI2022

Monitoring of Temperature Measurements for Different Flow Regimes in Water and Galinstan with Long Short-Term Memory Networks and Transfer Learning of Sensors

Temperature sensing is one of the most common measurements of a nuclear reactor monitoring system. The coolant fluid flow in a reactor core depends on the reactor power state. We investigated the monitoring and estimation of the thermocouple time series using machine learning for a range of flow regimes. Measurement data were obtained, in two separate experiments, in a flow loop filled with water and with liquid metal Galinstan. We developed long short-term memory (LSTM) recurrent neural networks (RNNs) for sensor predictions by training on the sensor’s own prior history, and transfer learning LSTM (TL-LSTM) by training on a correlated sensor’s prior history. Sensor cross-correlations were identified by calculating the Pearson correlation coefficient of the time series. The accuracy of LSTM and TL-LSTM predictions of temperature was studied as a function of Reynolds number (Re). The root-mean-square error (RMSE) for the test segment of time series of each sensor was shown to linearly increase with Re for both water and Galinstan fluids. Using linear correlations, we estimated the range of values of Re for which RMSE is smaller than the thermocouple measurement uncertainty. For both water and Galinstan fluids, we showed that both LSTM and TL-LSTM provide reliable estimations of temperature for typical flow regimes in a nuclear reactor. The LSTM runtime was shown to be substantially smaller than the data acquisition rate, which allows for performing estimation and validation of sensor measurements in real time.

Pantopoulou, Stella↗
DOE OSTI2024

Complexity-calibrated benchmarks for machine learning reveal when prediction algorithms succeed and mislead

Abstract Recurrent neural networks are used to forecast time series in finance, climate, language, and from many other domains. Reservoir computers are a particularly easily trainable form of recurrent neural network. Recently, a “next-generation” reservoir computer was introduced in which the memory trace involves only a finite number of previous symbols. We explore the inherent limitations of finite-past memory traces in this intriguing proposal. A lower bound from Fano’s inequality shows that, on highly non-Markovian processes generated by large probabilistic state machines, next-generation reservoir computers with reasonably long memory traces have an error probability that is at least $$\sim 60\%$$ ∼ 60 % higher than the minimal attainable error probability in predicting the next observation. More generally, it appears that popular recurrent neural networks fall far short of optimally predicting such complex processes. These results highlight the need for a new generation of optimized recurrent neural network architectures. Alongside this finding, we present concentration-of-measure results for randomly-generated but complex processes. One conclusion is that large probabilistic state machines—specifically, large $$\epsilon$$ ϵ -machines—are key to generating challenging and structurally-unbiased stimuli for ground-truthing recurrent neural network architectures.

97 MATHEMATICS AND COMPUTING↗