Gaussian process (GP) emulator has been used as a surrogate model for predicting force field and molecular potential, to overcome the computational bottleneck of ab initio molecular dynamics simulation. Integrating both atomic force and energy in predictions was found to be more accurate than using energy alone, yet it requires O(( NM) 3 ) computational operations for computing the likelihood function and making predictions, where N is the number of atoms and M is the number of simulated configurations in the training sample due to the inversion of a large covariance matrix. The high computational cost limits its applications to the simulation of small molecules. The computational challenge of using both gradient information and function values in GPs was recently noticed in machine learning communities, whereas conventional approximation methods may not work well. Here, we introduce a new approach, the atomized force field model, that integrates both force and energy in the emulator with many fewer computational operations. The drastic reduction in computation is achieved by utilizing the naturally sparse covariance structure that satisfies the constraints of the energy conservation and permutation symmetry of atoms. The efficient machine learning algorithm extends the limits of its applications on larger molecules under the same computational budget, with nearly no loss of predictive accuracy. Furthermore, our approach contains an uncertainty assessment of predictions of atomic forces and energies, useful for developing a sequential design over the chemical input space.
more »
« less
Reliable emulation of complex functionals by active learning with error control
A statistical emulator can be used as a surrogate of complex physics-based calculations to drastically reduce the computational cost. Its successful implementation hinges on an accurate representation of the nonlinear response surface with a high-dimensional input space. Conventional “space-filling” designs, including random sampling and Latin hypercube sampling, become inefficient as the dimensionality of the input variables increases, and the predictive accuracy of the emulator can degrade substantially for a test input distant from the training input set. To address this fundamental challenge, we develop a reliable emulator for predicting complex functionals by active learning with error control (ALEC). The algorithm is applicable to infinite-dimensional mapping with high-fidelity predictions and a controlled predictive error. The computational efficiency has been demonstrated by emulating the classical density functional theory (cDFT) calculations, a statistical-mechanical method widely used in modeling the equilibrium properties of complex molecular systems. We show that ALEC is much more accurate than conventional emulators based on the Gaussian processes with “space-filling” designs and alternative active learning methods. In addition, it is computationally more efficient than direct cDFT calculations. ALEC can be a reliable building block for emulating expensive functionals owing to its minimal computational cost, controllable predictive error, and fully automatic features.
more »
« less
- NSF-PAR ID:
- 10408464
- Date Published:
- Journal Name:
- The Journal of Chemical Physics
- Volume:
- 157
- Issue:
- 21
- ISSN:
- 0021-9606
- Page Range / eLocation ID:
- 214109
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Stochastic computing (SC) is a low-cost computational paradigm that has promising applications in digital filter design, image processing, and neural networks. Fundamental to these applications is the weighted addition operation, which is most often implemented by a multiplexer (mux) tree. Mux-based adders have very low area but typically require long bitstreams to reach practical accuracy thresholds when the number of summands is large. In this work, we first identify the main contributors to mux adder error. We then demonstrate with analysis and experiment that two new techniques, precise sampling and full correlation, can target and mitigate these error sources. Implementing these techniques in hardware leads to the design of CeMux (Correlation-enhanced Multiplexer), a stochastic mux adder that is significantly more accurate and uses much less area than traditional weighted adders. We compare CeMux to other SC and hybrid designs for an electrocardiogram filtering case study that employs a large digital filter. One major result is that CeMux is shown to be accurate even for large input sizes. CeMux's higher accuracy leads to a latency reduction of 4× to 16× over other designs. Furthermore, CeMux uses about 35% less area than existing designs, and we demonstrate that a small amount of accuracy can be traded for a further 50% reduction in area. Finally, we compare CeMux to a conventional binary design and we show that CeMux can achieve a 50% to 73% area reduction for similar power and latency as the conventional design but at a slightly higher level of error.more » « less
-
more » « less
For energy-assisted compression ignition (EACI) engine propulsion at high-altitude operating conditions using sustainable jet fuels with varying cetane numbers, it is essential to develop an efficient engine control system for robust and optimal operation. Control systems are typically trained using experimental data, which can be costly and time consuming to generate due to setup time of experiments, unforeseen delays/issues with manufacturing, mishaps/engine failures and the consequent repairs (which can take weeks), and errors in measurements. Computational fluid dynamics (CFD) simulations can overcome such burdens by complementing experiments with simulated data for control system training. Such simulations, however, can be computationally expensive. Existing data-driven machine learning (ML) models have shown promise for emulating the expensive CFD simulator, but encounter key limitations here due to the expensive nature of the training data and the range of differing combustion behaviors (e.g. misfires and partial/delayed ignition) observed at such broad operating conditions. We thus develop a novel physics-integrated emulator, called the Misfire-Integrated GP (MInt-GP), which integrates important auxiliary information on engine misfires within a Gaussian process surrogate model. With limited CFD training data, we show the MInt-GP model can yield reliable predictions of in-cylinder pressure evolution profiles and subsequent heat release profiles and engine CA50 predictions at a broad range of input conditions. We further demonstrate much better prediction capabilities of the MInt-GP at different combustion behaviors compared to existing data-driven ML models such as kriging and neural networks, while also observing up to 80 times computational speed-up over CFD, thus establishing its effectiveness as a tool to assist CFD for fast data generation in control system training.
-
Active learning with generalized sliced inverse regression for high-dimensional reliability analysisIt is computationally expensive to predict reliability using physical models at the design stage if many random input variables exist. This work introduces a dimension reduction technique based on generalized sliced inverse regression (GSIR) to mitigate the curse of dimensionality. The proposed high dimensional reliability method enables active learning to integrate GSIR, Gaussian Process (GP) modeling, and Importance Sampling (IS), resulting in an accurate reliability prediction at a reduced computational cost. The new method consists of three core steps, 1) identification of the importance sampling region, 2) dimension reduction by GSIR to produce a sufficient predictor, and 3) construction of a GP model for the true response with respect to the sufficient predictor in the reduced-dimension space. High accuracy and efficiency are achieved with active learning that is iteratively executed with the above three steps by adding new training points one by one in the region with a high chance of failure.more » « less
-
more » « less
Abstract Machine learning (ML) has been applied to space weather problems with increasing frequency in recent years, driven by an influx of in-situ measurements and a desire to improve modeling and forecasting capabilities throughout the field. Space weather originates from solar perturbations and is comprised of the resulting complex variations they cause within the numerous systems between the Sun and Earth. These systems are often tightly coupled and not well understood. This creates a need for skillful models with knowledge about the confidence of their predictions. One example of such a dynamical system highly impacted by space weather is the thermosphere, the neutral region of Earth’s upper atmosphere. Our inability to forecast it has severe repercussions in the context of satellite drag and computation of probability of collision between two space objects in low Earth orbit (LEO) for decision making in space operations. Even with (assumed) perfect forecast of model drivers, our incomplete knowledge of the system results in often inaccurate thermospheric neutral mass density predictions. Continuing efforts are being made to improve model accuracy, but density models rarely provide estimates of confidence in predictions. In this work, we propose two techniques to develop nonlinear ML regression models to predict thermospheric density while providing robust and reliable uncertainty estimates: Monte Carlo (MC) dropout and direct prediction of the probability distribution, both using the negative logarithm of predictive density (NLPD) loss function. We show the performance capabilities for models trained on both local and global datasets. We show that the NLPD loss provides similar results for both techniques but the direct probability distribution prediction method has a much lower computational cost. For the global model regressed on the Space Environment Technologies High Accuracy Satellite Drag Model (HASDM) density database, we achieve errors of approximately 11% on independent test data with well-calibrated uncertainty estimates. Using an in-situ CHAllenging Minisatellite Payload (CHAMP) density dataset, models developed using both techniques provide test error on the order of 13%. The CHAMP models—on validation and test data—are within 2% of perfect calibration for the twenty prediction intervals tested. We show that this model can also be used to obtain global density predictions with uncertainties at a given epoch.
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1063/5.0121805
