SEARCH · Engineering Papers
Results for “decision tree”
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.
Decision Tree Ensemble Machine Learning for Rapid QSTS Simulations.
Abstract not provided.
Evolving Decision Trees to Detect Anomalies in Recurrent ICS Networks.
Abstract not provided.
A decision tree algorithm for investigation of model biases related to dynamical cores and physical parameterizations: CESM/CAM EVALUATION BY DECISION TREES
An object-based evaluation method using a pattern recognition algorithm (i.e., classification trees) is applied to the simulated orographic precipitation for idealized experimental setups using the National Center of Atmospheric Research (NCAR) Community Atmosphere Model (CAM) with the finite volume (FV) and the Eulerian spectral transform dynamical cores with varying resolutions. Daily simulations were analyzed and three different types of precipitation features were identified by the classification tree algorithm. The statistical characteristics of these features (i.e., maximum value, mean value, and variance) were calculated to quantify the difference between the dynamical cores and changing resolutions. Even with the simple and smooth topography in the idealized setups, complexity in the precipitation fields simulated by the models develops quickly. The classification tree algorithm using objective thresholding successfully detected different types of precipitation features even as the complexity of the precipitation field increased. The results show that the complexity and the bias introduced in small-scale phenomena due to the spectral transform method of CAM Eulerian spectral dynamical core is prominent, and is an important reason for its dissimilarity from the FV dynamical core. The resolvable scales, both in horizontal and vertical dimensions, have significant effect on the simulation of precipitation. The results of this study also suggest that an efficient and informative study about the biases produced by GCMs should involve daily (or even hourly) output (rather than monthly mean) analysis over local scales.
Optimal decision trees for categorical data via integer programming
Decision trees have been a very popular class of predictive models for decades due to their interpretability and good performance on categorical features. However, they are not always robust and tend to overfit the data. Additionally, if allowed to grow large, they lose interpretability. In this paper, we present a mixed integer programming formulation to construct optimal decision trees of a prespecified size. We take the special structure of categorical features into account and allow combinatorial decisions (based on subsets of values of features) at each node. Our approach can also handle numerical features via thresholding. Here we show that very good accuracy can be achieved with small trees using moderately-sized training sets. The optimization problems we solve are tractable with modern solvers.
Properly Learning Decision Trees in almost Polynomial Time
We give an n O (log log n ) -time membership query algorithm for properly and agnostically learning decision trees under the uniform distribution over { ± 1} n . Even in the realizable setting, the previous fastest runtime was n O (log n ) , a consequence of a classic algorithm of Ehrenfeucht and Haussler. Our algorithm shares similarities with practical heuristics for learning decision trees, which we augment with additional ideas to circumvent known lower bounds against these heuristics. To analyze our algorithm, we prove a new structural result for decision trees that strengthens a theorem of O’Donnell, Saks, Schramm, and Servedio. While the OSSS theorem says that every decision tree has an influential variable, we show how every decision tree can be “pruned” so that every variable in the resulting tree is influential.
Linear model decision trees as surrogates in optimization of engineering applications
Machine learning models are promising as surrogates in optimization when replacing difficult to solve equations or black-box type models. This work demonstrates the viability of linear model decision trees as piecewise-linear surrogates in decision-making problems. Linear model decision trees can be represented exactly in mixed-integer linear programming (MILP) and mixed-integer quadratic constrained programming (MIQCP) formulations. Furthermore, they can represent discontinuous functions, bringing advantages over neural networks in some cases. We present several formulations using transformations from Generalized Disjunctive Programming (GDP) formulations and modifications of MILP formulations for gradient boosted decision trees (GBDT). We then compare the computational performance of these different MILP and MIQCP representations in an optimization problem and illustrate their use on engineering applications. Importantly, we observe faster solution times for optimization problems with linear model decision tree surrogates when compared with GBDT surrogates using the Optimization and Machine Learning Toolkit (OMLT).
Enhanced Oblique Decision Tree Enabled Policy Extraction for Deep Reinforcement Learning in Power System Emergency Control
Deep reinforcement learning (DRL) algorithms have successfully solved many challenging problems in various power system control scenarios. However, their decision-making process is usually regarded as black-boxes. Furthermore, how DRL models interact with human intelligence remains an open problem. Thus, this paper proposes a policy extraction framework to extract a complex DRL model into an explainable policy. This framework includes three parts: 1) DRL training and data generation. We train an agent for a specific control task and generate data, which contains the control policy of the agent. 2) Policy extraction. We propose an information gain rate based weighted oblique decision tree (IGR-WODT) for DRL policy extraction. 3) Policy evaluation. We define three metrics to evaluate the performance of the proposed approach. A case study for the under-voltage load shedding problem shows that the IGR-WODT presents a performance enhancement compared with DRL, weighted oblique decision tree, and univariate decision tree. The proposed policy extraction method could provide an intuitive explanation of the neural network decision-making process to the dispatchers when making final decisions on power grid operation. Also, the resulted rule-based controller could replace the deep neural network-based controller in many field edge devices with limited computing resources, providing comparable performance.
Office of Legacy Management Decision Tree for Solar Photovoltaic Projects - 13317
To support consideration of renewable energy power development as a land reuse option, the DOE Office of Legacy Management (LM) and the National Renewable Energy Laboratory (NREL) established a partnership to conduct an assessment of wind and solar renewable energy resources on LM lands. From a solar capacity perspective, the larger sites in the western United States present opportunities for constructing solar photovoltaic (PV) projects. A detailed analysis and preliminary plan was developed for three large sites in New Mexico, assessing the costs, the conceptual layout of a PV system, and the electric utility interconnection process. As a result of the study, a 1,214-hectare (3,000-acre) site near Grants, New Mexico, was chosen for further study. The state incentives, utility connection process, and transmission line capacity were key factors in assessing the feasibility of the project. LM's Durango, Colorado, Disposal Site was also chosen for consideration because the uranium mill tailings disposal cell is on a hillside facing south, transmission lines cross the property, and the community was very supportive of the project. LM worked with the regulators to demonstrate that the disposal cell's long-term performance would not be impacted by the installation of a PV solar system. A number of LM-unique issues were resolved in making the site available for a private party to lease a portion of the site for a solar PV project. A lease was awarded in September 2012. Using a solar decision tree that was developed and launched by the EPA and NREL, LM has modified and expanded the decision tree structure to address the unique aspects and challenges faced by LM on its multiple sites. The LM solar decision tree covers factors such as land ownership, usable acreage, financial viability of the project, stakeholder involvement, and transmission line capacity. As additional sites are transferred to LM in the future, the decision tree will assist in determining whether a solar PV project is feasible on the new sites. (authors)
A Decision Tree Based Machine Learning Algorithm for Rate Estimation
A subgraph isomorphic decision tree algorithm was developed for estimating chemical reaction rates. The algorithm essentially associates differences in rate estimation with differences in sub-structures on the reactant or reactants, and the tree splits reactions into groups with similar rates by identifying the substructural differences that affect rate estimation the most. The algorithm was trained on rates from the Reaction Mechanism Generator (RMG) database and compared with a much more traditional rate rule estimator based on the same data. Compared to the previous RMG rate rules, the prediction accuracy of the new algorithm, measured by median absolute error (MAE) factor, root mean squared error (RMSE) factor, and 2-sigma error factor, is consistently improved by a factor of two. Uncertainty estimation was also developed for the decision trees based on fitting a lognormal distribution to the leave-one-out errors at each node. Improvements in accuracy introduced by the algorithm should significantly reduce the amount of effort required to refine models and the need for additional quantum calculations and thus reduce the amount of time and cost required to create and refine biofuel models.
Fast inference of Boosted Decision Trees in FPGAs for particle physics
We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extremely low latency. With a typical latency less than 100 ns, this solution is suitable for FPGA-based real-time processing, such as in the Level-1 Trigger system of a collider experiment. These developments open up prospects for physicists to deploy BDTs in FPGAs for identifying the origin of jets, better reconstructing the energies of muons, and enabling better selection of rare signal processes.
PySIDT: Subgraph Isomorphic Decision Trees for Molecular Property Prediction
Accurate molecular property prediction is important across all fields of chemistry. Deep neural networks (DNNs) have become increasingly popular due to their ability to train automatically, avoiding the incredibly tedious process of constructing and extending traditional property estimation schemes. However, DNNs require large amounts of training data, are challenging to interpret, require large amounts of memory to load even during inference, and have severe difficulties incorporating qualitative chemical knowledge, which are often desired for molecular property prediction tasks. Here, in this study, we present PySIDT (https://github.com/zadorlab/PySIDT), a software for training and running inference on Subgraph Isomorphic Decision Trees (SIDTs). SIDTs are graph-based decision trees made of nodes associated with molecular substructures. Inference is done by descending target molecular structures down the decision tree to nodes with matching subgraph isomorphic substructures and making predictions based on the final (most specific) nodes matched. SIDTs scale down well to dataset sizes much smaller than is feasible for DNNs. As trees of molecular substructures, SIDTs are inherently readable and easy to visualize, making them easy to analyze. They are also straightforward to extend and retrain, facilitate uncertainty estimation, and enable easy integration of expert knowledge. We demonstrate the SIDT approach discussing its application to a diverse range of molecular prediction tasks: rate coefficient estimation, diffusion coefficient estimation, thermochemistry estimation, transition state bond stretch prediction, p K a prediction, stability of molecular structures, stability of surface structures, and prediction of surface lateral interaction energetics. Additionally, we demonstrate the power of the SIDT algorithms in two direct learning curve vanilla comparisons with the popular DNN-based software Chemprop and the popular gradient boosted trees-based software XGBoost on enthalpy of formation and rate coefficient prediction tasks. In particular, in the enthalpy of formation case, vanilla PySIDT is able to outperform vanilla Chemprop and XGBoost across the full range of training/validation set sizes out to 11,560 data points.
Disentangling error structures of precipitation datasets using decision trees
Characterizing error structures in precipitation products not only facilitates their proper applications for scientific and practical purposes but also helps improve their retrieval algorithms and processing methods. Despite the fact that multiple precipitation products have been assessed in the literature, factors that affect their error structures remain inadequately addressed. By interpreting 60 binary decision trees, this study disentangles the error characteristics of precipitation products in terms of their spatiotemporal patterns and geographical factors. Three independent precipitation products - two satellite-based and one reanalysis datasets: the Integrated Multi-satellitE Retrievals for GPM (Global Precipitation Measurement) late run (IMERG-L), Soil Moisture to Rain-Advanced SCATterometer (SM2RAIN-ASCAT), and the Modern-Era Retrospective analysis for Research and Applications, Version 2 uncorrected precipitation output (MERRA2-UC), are evaluated across the contiguous United States from 2010 to 2019. Here, the ground-based Stage IV precipitation dataset is used as the ground truth. Results indicate that the MERRA2-UC outperforms the IMERG-L and SM2RAIN-ASCAT with higher accuracy and more stable interannual patterns for the analysis period. Decision trees cross-assess three spatiotemporal factors and find that the underestimation of MERRA2-UC occurs in the east of the Rocky Mountains, and SM2RAIN-ASCAT underestimates precipitation over high latitudes, especially in winter. Additionally, the decision tree method ascribes system errors to nine different geographical characteristics, of which the distance to the coast, soil type, and DEM are the three dominant features. On the other hand, the land cover type, topography position index, and aspect are three relatively weak factors.
Biomass for Carbon Removal and Storage (BiCRS) Counterfactual Decision Tree
Counterfactual is the term used to describe a "business-as-usual" scenario which used as a baseline to compare against a new project, allowing the calculation of net impacts for a life cycle analysis (LCA). The choice of counterfactual is critical for determining the results from LCA and must be carefully justified to ensure a fair and accurate comparison. Using forest residues as an example, this decision tree illustrates decision points to be considered for sustainable biomass sourcing and provides a framework for estimating the carbon emissions or storage under the "business-as-usual” scenarios for biomass otherwise destined for use in Biomass for Carbon Removal and Storage (BiCRS) projects.
Nanosecond machine learning regression with deep boosted decision trees in FPGA for high energy physics
We present a novel application of the machine learning / artificial intelligence method called boosted decision trees to estimate physical quantities on field programmable gate arrays (FPGA). The software package fwXmachina features a new architecture called parallel decision paths that allows for deep decision trees with arbitrary number of input variables. It also features a new optimization scheme to use different numbers of bits for each input variable, which produces optimal physics results and ultraefficient FPGA resource utilization. Problems in high energy physics of proton collisions at the Large Hadron Collider (LHC) are considered. Estimation of missing transverse momentum (E T miss ) at the first level trigger system at the High Luminosity LHC (HL-LHC) experiments, with a simplified detector modeled by Delphes, is used to benchmark and characterize the firmware performance. The firmware implementation with a maximum depth of up to 10 using eight input variables of 16-bit precision gives a latency value of $\mathcal{O}$(10) ns, independent of the clock speed, and $\mathcal{O}$(0.1)% of the available FPGA resources without using digital signal processors.
Using Boosted Decision Trees to Select High Quality Measurements in the Mu2e Experiment at Fermilab
This thesis presents the implementation and evaluation of a Boosted Decision Tree (BDT) model to improve the selection of high-quality track measurements in the Mu2e experiment at Fermilab. The Mu2e experiment is a high-energy physics experiments seeking to observe a rare theoretical physics process known as Charged Lepton Flavor Violation. A significant challenge faced by the Mu2e experiment are so-called background events, which are events whose data mimics that of the rare physics process the experiment seeks to observe. Without a mechanism to reduce background, it would be impossible to know whether Charged Lepton Flavor Violation occurred or not. To this end, high-quality track measurements must be distinguished from low-quality track measurements. A track can be conceived of as the reconstructed path of a particle that traveled through the Mu2e detector. In addition to other data, data about such tracks is stored using a C++-based framework, specific to the domain of high-energy physics, known as ROOT. A boosted decision tree model was trained using ROOT’s Toolkit For Multivariate Analysis by leveraging variables ancillary to track quality. In evaluation, the BDT achieves a ROC-AUC of 0.927 in discriminating good-quality tracks from poor-quality tracks. Such a score is indicative of both strong discrimination and strong generalization. Subsequently, it is shown that applying a BDT-based quality cut to the distribution of particle momenta significantly enhances the signal-to-background distinction for signal electrons, paving the way for improved sensitivity to Charged Lepton Flavor Violation.
Decision Tree for Variable Selection vs. Impact on Durability for Biomass and Biochar Burial Pathways [Slides]
Quantifying durability for lower-TRL BiCRS pathways has been challenging as limited data are available from real-world projects and long-term experiments, resulting in an overall lack of scientific consensus. We develop a decision tree that aims to summarize the current scientific understanding and state-of-the-art project experience. The decision tree can be used to (1) guide the selection of key variables and evaluate their relative impact on durability, (2) identify data and knowledge gaps for future research.
Development of heavy-duty vehicle representative driving cycles via decision tree regression
Previously, researchers who developed representative driving cycles mainly focused on light-duty vehicles and only considered vehicle speed and related derivations. In this paper, we propose a novel approach to develop representative cycles for heavy-duty vehicles. By implementing decision tree regression (DTR) to the Fleet DNA on-road vehicle data, a broader set of metrics, such as engine power and fuel consumption, can be used for more robust cycle development. Additionally, the influence of each metric on the regression target is also accounted for by a weighted number derived through the DTR to enhance the representativenss of the developed cycle. As case studies, we applied the proposed method to five heavy-duty vocations (drayage, long haul, regional haul, local delivery, and transit bus) and derived the most representative cycle, as well as four extreme cycles (maximal energy consumption, maximal power-weighted work, maximal fraction of high speed, and minimal fuel economy) to advance the related alternative powertrain design.