Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ROSES”

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 1,459 records · Page 81
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Optimizing deep learning hyper-parameters through an evolutionary algorithm. In: MLHPC '15 Proceedings of the Workshop on Machine Learning in High-Performance Computing Environments, Article No. 4

There has been a recent surge of success in utilizing Deep Learning (DL) in imaging and speech applications for its relatively automatic feature generation and, in particular for convolutional neural networks (CNNs), high accuracy classification abilities. While these models learn their parameters through data-driven methods, model selection (as architecture construction) through hyper-parameter choices remains a tedious and highly intuition driven task. To address this, Multi-node Evolutionary Neural Networks for Deep Learning (MENNDL) is proposed as a method for automating network selection on computational clusters through hyper-parameter optimization performed via genetic algorithms.

Young, Steven R.↗
DOE OSTI2017

A Programming Framework for Neuromorphic Systems with Emerging Technologies

Neuromorphic computing is a promising post-Moore's law era technology. A wide variety of neuromorphic computer (NC) architectures have emerged in recent years, ranging from traditional fully digital CMOS to nanoscale implementations with novel, beyond CMOS components. There are already major questions associated with how we are going to program and use NCs simply because of how radically different their architecture is as compared with the von Neumann architecture. When coupled with the implementations using emerging device technologies, which add additional issues associated with programming devices, it is clear that we must define a new way to program and develop for NC devices. In this work, we discuss a programming framework for NC devices implemented with emerging technologies. We discuss how we have applied this framework to program a NC system implemented with metal oxide memristors. We utilize the framework to develop two applications for the memristive NC device: a simple multiplexer and a simple control task (the cart-pole problem). Finally, we discuss how this framework can be extended to NC systems implemented with a variety of novel device components and materials.

Schuman, Catherine↗
DOE OSTI2018

A Study of Complex Deep Learning Networks on High-Performance, Neuromorphic, and Quantum Computers

Current deep learning approaches have been very successful using convolutional neural networks trained on large graphical-processing-unit-based computers. Three limitations of this approach are that (1) they are based on a simple layered network topology, i.e., highly connected layers, without intra-layer connections; (2) the networks are manually configured to achieve optimal results, and (3) the implementation of the network model is expensive in both cost and power. In this paper, we evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation. We use the MNIST dataset for our experiment, due to input size limitations of current quantum computers. Our results show the feasibility of using the three architectures in tandem to address the above deep learning limitations. Finally, we show that a quantum computer can find high quality values of intra-layer connection weights in a tractable time as the complexity of the network increases, a high performance computer can find optimal layer-based topologies, and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Evaluating online-learning in memristive neuromorphic circuits

In this work we present an implementation of spike-timing-dependent plasticity (STDP) in both a high level simulation and at a circuit level. It is verified that the high level simulation captures the behavior of the circuit implementation. We use the simulation to assess the effectiveness of STDP for online-learning, and find that STDP enables networks to improve performance online after training.

Wyer, Austin↗
DOE OSTI2018

Energy and Area Efficiency in Neuromorphic Computing for Resource Constrained Devices

Resource constrained devices are the building blocks of the internet of things (IoT) era. Since the idea behind IoT is to develop an interconnected environment where the devices are tiny enough to operate with limited resources, several control systems have been built to maintain low energy and area consumption while operating as IoT edge devices. Several researchers have begun work on implementing control systems built from resource constrained devices using machine learning. However, there are many ways such devices can achieve lower power consumption and area utilization while maximizing application efficiency. Spiky neuromorphic computing (SNC) is an emerging paradigm that can be leveraged in resource constrained devices for several emerging applications. While delivering the benefits of machine learning, SNC also helps minimize power consumption. For example, low energy memory devices (memristors) are often used to achieve low power operation and also help in reducing system area. In total, we anticipate SNC will provide computational efficiency approaching that of deep learning while using low power, resource constrained devices.

Chakma, Gangotree↗
DOE OSTI2018

A Comparison of Neuromorphic Classification Tasks

A variety of neural network models and machine learning techniques have arisen over the past decade, and their successes with image classification have been stunning. With other classification tasks, selecting and configuring a neural network solution is not straightforward. In this paper, we evaluate and compare a variety of neural network models, trained by a variety of machine learning techniques, on a variety of classification tasks. While Deep Learning typically exhibits the best classification accuracy, we note the promise of Reservoir Computing, and evolutionary optimization on spiking neural networks. In many cases, these technologies perform as well as, or better than Deep Learning, and the resulting networks are much smaller than their Deep Learning counterparts.

Reynolds, John↗
DOE OSTI2019

Shortest Path and Neighborhood Subgraph Extraction on a Spiking Memristive Neuromorphic Implementation

Spiking neuromorphic computers (SNCs) are promising as a post Moore's law technology partly because of their potential for very low power computation. SNCs have primarily been demonstrated on machine learning and neural network applications, but they can also be used for applications beyond machine learning that can leverage SNC properties such as massively parallel computation and collocated processing and memory. Here, we demonstrate two graph problems (shortest path and neighborhood subgraph extraction) that can be solved using SNCs. We discuss the approach for mapping these applications to an SNC. We also estimate the performance of a memristive SNC for these applications on three real-world graphs.

Schuman, Catherine↗
DOE OSTI2021

Design of a Robust Memristive Spiking Neuromorphic System with Unsupervised Learning in Hardware

Spiking neural networks (SNN) offer a power efficient, biologically plausible learning paradigm by encoding information into spikes. The discovery of the memristor has accelerated the progress of spiking neuromorphic systems, as the intrinsic plasticity of the device makes it an ideal candidate to mimic a biological synapse. Despite providing a nanoscale form factor, non-volatility, and low-power operation, memristors suffer from device-level non-idealities, which impact system-level performance. To address these issues, this article presents a memristive crossbar-based neuromorphic system using unsupervised learning with twin-memristor synapses, fully digital pulse width modulated spike-timing-dependent plasticity, and homeostasis neurons. Additionally, the implemented single-layer SNN was applied to a pattern-recognition task of classifying handwritten-digits. The performance of the system was analyzed by varying design parameters such as number of training epochs, neurons, and capacitors. Furthermore, the impact of memristor device non-idealities, such as device-switching mismatch, aging, failure, and process variations, were investigated and the resilience of the proposed system was demonstrated.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Accurate and Accelerated Neuromorphic Network Design Leveraging A Bayesian Hyperparameter Pareto Optimization Approach

Neuromorphic systems allow for extremely efficient hardware implementations for neural networks (NNs). In recent years, several algorithms have been presented to train spiking NNs (SNNs) for neuromorphic hardware. However, SNNs often provide lower accuracy than their artificial NNs (ANNs) counterparts or require computationally expensive and slow training/inference methods. To close this gap, designers typically rely on reconfiguring SNNs through adjustments in the neuron/synapse model or training algorithm itself. Nevertheless, these steps incur significant design time, while still lacking the desired improvement in terms of training/inference times (latency). Designing SNNs that can mimic the accuracy of ANNs with reasonable training times is an exigent challenge in neuromorphic computing. In this work, we present an alternative approach that looks at such designs as an optimization problem rather than algorithm or architecture redesign. We develop a versatile multiobjective hyperparameter optimization (HPO) for automatically tuning HPs of two state-of-the-art SNN training algorithms, SLAYER and HYBRID. We emphasize that, to the best of our knowledge, this is the first work trying to improve SNNs’ computational efficiency, accuracy, and training time using an efficient HPO. We demonstrate significant performance improvements for SNNs on several datasets without the need to redesign or invent new training algorithms/architectures. Our approach results in more accurate networks with lower latency and, in turn, higher energy efficiency than previous implementations. In particular, we demonstrate improvement in accuracy and more than 5× reduction in the training/inference time for the SLAYER algorithm on the DVS Gesture dataset. In the case of HYBRID, we demonstrate 30% reduction in timesteps while surpassing the accuracy of the state-of-the-art networks on CIFAR10. Further, our analysis suggests that even a seemingly minor change in HPs could change the accuracy by 5 - 6×.

Parsa, Maryam↗
DOE OSTI2022

A Review of Bayesian Networks for Spatial Data

We report Bayesian networks are a popular class of multivariate probabilistic models as they allow for the translation of prior beliefs about conditional dependencies between variables to be easily encoded into their model structure. Due to their widespread usage, they are often applied to spatial data for inferring properties of the systems under study and also generating predictions for how these systems may behave in the future. We review published research on methodologies for representing spatial data with Bayesian networks and also summarize the application areas for which Bayesian networks are employed in the modeling of spatial data. We find that a wide variety of perspectives are taken, including a GIS-centric focus on efficiently generating geospatial predictions, a statistical focus on rigorously constructing graphical models controlling for spatial correlation, as well as a range of problem-specific heuristics for mitigating the effects of spatial correlation and dependency arising in spatial data analysis. Special attention is also paid to potential future directions for integration of Bayesian networks with spatial processes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Fragile Earth: AI for Climate Mitigation, Adaptation, and Environmental Justice

The Fragile EarthWorkshop is a recurring event that gathers the research community to find and explore howdata science can measure and progress climate and social issues, following the framework of the United Nations Sustainable Development Goals (SDGs).Fragile Earth 2022: AI for Climate Mitigation, Adaptation, and Environmental Justice is a workshop taking place as part of the ACM's KDD 2022 Conference on research in knowledge discovery and data mining and their applications. The dates for the Conference are August 14-18, 2022.

Abe, Naoki↗
DOE OSTI2021

Electrochemistry of the NaI-AlBr 3 Molten Salt System: A Redox-Active, Low-Temperature Molten Salt Electrolyte

NaI-AlBr 3 is a very appealing low melting temperature (<100 °C), salt system for use as an electrochemically-active electrolyte. This system was investigated for its electrochemical and physical properties with focus to energy storage considerations. A simple phase diagram was generated; at >100 °C, lower NaI concentrations had two partially miscible liquid phases, while higher NaI concentrations had solid particles. Considering the fully molten regime, electrical conductivities were evaluated over 5–25 mol% NaI and 110 °C–140 °C. Conductivities of 6.8–38.9 mS cm −1 were observed, increasing with temperature and NaI concentration. Effective diffusion coefficients of the I − /I 3 − redox species were found to decrease with both increasing NaI concentration and increasing applied potential. Regardless, oxidation current density at 3.6 V vs Na/Na + was observed to increase with increasing NaI concentration over 5–25 mol%. Finally, the critical interface between the molten salt electrolyte and electrode materials was found to significantly affect reaction kinetics. When carbon was used instead of tungsten, an adsorbed species, most likely I 2 , blocked surface sites and significantly decreased current densities at high potentials. This study shows the NaI-AlBr 3 system offers an attractive, low-temperature molten salt electrolyte that could be useful to many applied systems, though composition and electrode material must be considered.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Electrochemical Modeling of Iodide Oxidation in Metal-Halide Molten Salts

The oxidation of iodide in NaI-AlBr 3 , NaI-AlCl 3 , and NaI-GaCl 3 molten salts was analyzed using simulation software to extract relevant kinetic parameters. The experimental oxidation potentials were ordered AlCl 3 < AlBr 3 < GaCl 3 , with higher oxidation potentials correlating with softer Lewis acidity of the metal halide. An iodide oxidation and metal halide speciation model was developed and simulated to fit the electrochemical response, enabling determination of electrochemical charge transfer parameters and chemical equilibrium constants. NaI-AlBr 3 displayed the fastest electron transfer rates yet showed the lowest current densities. All salts revealed smaller than expected current densities, explained by equilibrium between various species, where some are not electrochemically active at the studied potentials. These equilibrium reactions are due to the various metal halide species, controlling the reactant concentration of iodide and the resultant current. We hypothesize the electrochemically active iodide species, present as a metal halide monomer (MX 3 I − ), is decreased dramatically from the expected concentration, sequestered as a more stable metal halide dimer species (M 2 X 6 I − ) with a higher oxidation potential. Traditional Tafel analysis of the experimental data supports the validity of the simulations. These results increase understanding of iodide oxidation in low-temperature Lewis acidic molten salts and inform task-specific molten salt design.

Lee, Rose Y. (ORCID:0000000226587220)↗
DOE OSTI2023

Electrode Blocking Due to Redox Reactions in Aluminum Chloride-Sodium Iodide Molten Salts

Iodide redox reactions in molten NaI/AlCl 3 are shown to generate surface-blocking films, which may limit the useful cycling rates and energy densities of molten sodium batteries below 150 °C. An experimental investigation of electrode interfacial stability at 110 °C reveals the source of the reaction rate limitations. Electrochemical experiments in a 3-electrode configuration confirm an increase of resistance on the electrode surface after oxidation or reduction current is passed. Using chronopotentiometry, chronoamperometry, cyclic voltammetry, and electrochemical impedance spectroscopy, the film formation is shown to depend on the electrode material (W, Mo, Ta, or glassy carbon), as well as the Lewis acidity and molar ratio of I − /I 3 − in the molten salt electrolytes. These factors impact the amount of charge that can be passed at a given current density prior to developing excessive overpotential due to film formation that blocks the electrode surface. The results presented here guide the design and use of iodide-based molten salt electrolytes and electrode materials for grid scale battery applications.

25 ENERGY STORAGE↗
DOE OSTI2014

Raman microscopy of lithium-manganese-rich transition metal oxide cathodes

Lithium-rich and manganese-rich (LMR) layered transition metal (TM) oxide composites with general formula xLi 2 MnO 3 ·(1-x)LiMO 2 (M = Ni, Co, Mn) are promising cathode candidates for high energy density lithium ion batteries. Lithium-manganese-rich TM oxides crystallize as a nanocomposite layered phase whose structure further evolves with electrochemical cycling. Raman spectroscopy is a powerful tool to monitor the crystal chemistry and correlate phase changes with electrochemical behavior. While several groups have reported Raman spectra of lithium rich TM oxides, the data show considerable variability in terms of both the vibrational features observed and their interpretation. In this paper, Raman microscopy is used to investigate lithium-rich and manganese-rich TM cathodes as a function of voltage and electrochemical cycling at various temperatures. No growth of a spinel phase is observed within the cycling conditions. However, analysis of the Raman spectra does indicate the structure of LMR-NMC deviates significantly from an ideal layered phase. Finally, the results also highlight the importance of using low laser power and large sample sizes to obtain consistent data sets.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗