Search for: All records

Abstract Coarse graining techniques play an essential role in accelerating molecular simulations of systems with large length and time scales. Theoretically grounded bottom-up models are appealing due to their thermodynamic consistency with the underlying all-atom models. In this direction, machine learning approaches hold great promise to fitting complex many-body data. However, training models may require collection of large amounts of expensive data. Moreover, quantifying trained model accuracy is challenging, especially in cases of non-trivial free energy configurations, where training data may be sparse. We demonstrate a path towards uncertainty-aware models of coarse grained free energy surfaces. Specifically, we show that principled Bayesian model uncertainty allows for efficient data collection through an on-the-fly active learning framework and opens the possibility of adaptive transfer of models across different chemical systems. Uncertainties also characterize models’ accuracy of free energy predictions, even when training is performed only on forces. This work helps pave the way towards efficient autonomous training of reliable and uncertainty aware many-body machine learned coarse grain models. 

The rapid development and large body of literature on machine learning interatomic potentials (MLIPs) can make it difficult to know how to proceed for researchers who are not experts but wish to use these tools. The spirit of this review is to help such researchers by serving as a practical, accessible guide to the state-of-the-art in MLIPs. This review paper covers a broad range of topics related to MLIPs, including (i) central aspects of how and why MLIPs are enablers of many exciting advancements in molecular modeling, (ii) the main underpinnings of different types of MLIPs, including their basic structure and formalism, (iii) the potentially transformative impact of universal MLIPs for both organic and inorganic systems, including an overview of the most recent advances, capabilities, downsides, and potential applications of this nascent class of MLIPs, (iv) a practical guide for estimating and understanding the execution speed of MLIPs, including guidance for users based on hardware availability, type of MLIP used, and prospective simulation size and time, (v) a manual for what MLIP a user should choose for a given application by considering hardware resources, speed requirements, energy and force accuracy requirements, as well as guidance for choosing pre-trained potentials or fitting a new potential from scratch, (vi) discussion around MLIP infrastructure, including sources of training data, pre-trained potentials, and hardware resources for training, (vii) summary of some key limitations of present MLIPs and current approaches to mitigate such limitations, including methods of including long-range interactions, handling magnetic systems, and treatment of excited states, and finally (viii) we finish with some more speculative thoughts on what the future holds for the development and application of MLIPs over the next 3–10+ years. 

Deep learning has emerged as a promising paradigm to give access to highly accurate predictions of molecular and material properties. A common short-coming shared by current approaches, however, is that neural networks only give point estimates of their predictions and do not come with predictive uncertainties associated with these estimates. Existing uncertainty quantification efforts have primarily leveraged the standard deviation of predictions across an ensemble of independently trained neural networks. This incurs a large computational overhead in both training and prediction, resulting in order-of-magnitude more expensive predictions. Here, we propose a method to estimate the predictive uncertainty based on a single neural network without the need for an ensemble. This allows us to obtain uncertainty estimates with virtually no additional computational overhead over standard training and inference. We demonstrate that the quality of the uncertainty estimates matches those obtained from deep ensembles. We further examine the uncertainty estimates of our methods and deep ensembles across the configuration space of our test system and compare the uncertainties to the potential energy surface. Finally, we study the efficacy of the method in an active learning setting and find the results to match an ensemble-based strategy at order-of-magnitude reduced computational cost. 

The next-generation semiconductors and devices, such as halide perovskites and flexible electronics, are extremely sensitive to water, thus demanding highly effective protection that not only seals out water in all forms (vapor, droplet, and ice), but simultaneously provides mechanical flexibility, durability, transparency, and self-cleaning. Although various solid-state encapsulation methods have been developed, no strategy is available that can fully meet all the above requirements. Here, we report a bioinspired liquid-based encapsulation strategy that offers protection from water without sacrificing the operational properties of the encapsulated materials. Using halide perovskite as a model system, we show that damage to the perovskite from exposure to water is drastically reduced when it is coated by a polymer matrix with infused hydrophobic oil. With a combination of experimental and simulation studies, we elucidated the fundamental transport mechanisms of ultralow water transmission rate that stem from the ability of the infused liquid to fill-in and reduce defects in the coating layer, thus eliminating the low-energy diffusion pathways, and to cause water molecules to diffuse as clusters, which act together as an excellent water permeation barrier. Importantly, the presence of the liquid, as the central component in this encapsulation method provides a unique possibility of reversing the water transport direction; therefore, the lifetime of enclosed water-sensitive materials could be significantly extended via replenishing the hydrophobic oils regularly. We show that the liquid encapsulation platform presented here has high potential in providing not only water protection of the functional device but also flexibility, optical transparency, and self-healing of the coating layer, which are critical for a variety of applications, such as in perovskite solar cells and bioelectronics. 

Electronic devicesforrecording neuralactivityinthe nervoussyste m needto bescalableacrosslargespatialandte mporalscales whilealso providing millisecondandsingle-cellspatiote mporalresolution. H o w e v e r, e xi s ti n g hi g h- r e s ol u ti o n n e u r al r e c o r di n g d e vi c e s c a n n o t achievesi multaneousscalability on bothspatialandte mporallevels due toatrade-offbetweensensordensityand mechanicalflexibility. Here weintroduceathree-di mensional(3D)stackingi mplantableelectronic platfor m,basedonperfluorinateddielectricelasto mersandtissue-levelsoft multilayerelectrodes,thatenablesspatiote mporallyscalablesingle-cell neuralelectrophysiologyinthenervoussyste m. Ourelasto mersexhibit stable dielectric perfor mancefor overayearin physiologicalsolutions andare10,000ti messofterthanconventional plastic dielectrics. By leveragingthese uniquecharacteristics we developthe packaging of lithographednano metre-thickelectrodearraysina3Dconfiguration with across-sectionaldensityof7.6electrodesper100μ m2.Theresulting3D integrated multilayersoftelectrodearrayretainstissue-levelflexibility, reducingchronici m muneresponsesin mouse neuraltissues,and de monstratestheabilitytoreliablytrackelectricalactivityinthe mouse brain orspinalcord over months without disruptingani mal behaviour. 

Abstract Machine learning interatomic force fields are promising for combining high computational efficiency and accuracy in modeling quantum interactions and simulating atomistic dynamics. Active learning methods have been recently developed to train force fields efficiently and automatically. Among them, Bayesian active learning utilizes principled uncertainty quantification to make data acquisition decisions. In this work, we present a general Bayesian active learning workflow, where the force field is constructed from a sparse Gaussian process regression model based on atomic cluster expansion descriptors. To circumvent the high computational cost of the sparse Gaussian process uncertainty calculation, we formulate a high-performance approximate mapping of the uncertainty and demonstrate a speedup of several orders of magnitude. We demonstrate the autonomous active learning workflow by training a Bayesian force field model for silicon carbide (SiC) polymorphs in only a few days of computer time and show that pressure-induced phase transformations are accurately captured. The resulting model exhibits close agreement with both ab initio calculations and experimental measurements, and outperforms existing empirical models on vibrational and thermal properties. The active learning workflow readily generalizes to a wide range of material systems and accelerates their computational understanding. 

Here we provide an in-depth structural characterization of the amorphous ionic glasses LiPON and LiSiPON with high Li content. Based on ab-initio molecular dynamics simulations, the structure of these materials is an inverted structure with either isolated polyanion tetrahedra or polyanion dimers suspended in a Li+ matrix. Based on neutron scattering data, this type of inverted structure leads to a significant amount of medium-range ordering in the structure, as demonstrated by two sharp diffraction peaks and a periodic structural oscillation in the density function G(r). On a local scale, adding N and Si increases the number of anion bridges and polyanion dimer structures, leading to higher ionic conductivity. In the medium range ordering, the addition of Si leads to more disorder in the polyanion substructure but a significant increase in the ordering of the O substructure. Finally, we demonstrate that this inverted structure with medium range ordering results in a glassy material that is both mechanically stiff and ductile on the nanoscale. 

Abstract Atomistic modeling of chemically reactive systems has so far relied on either expensive ab initio methods or bond-order force fields requiring arduous parametrization. Here, we describe a Bayesian active learning framework for autonomous “on-the-fly” training of fast and accurate reactive many-body force fields during molecular dynamics simulations. At each time-step, predictive uncertainties of a sparse Gaussian process are evaluated to automatically determine whether additional ab initio training data are needed. We introduce a general method for mapping trained kernel models onto equivalent polynomial models whose prediction cost is much lower and independent of the training set size. As a demonstration, we perform direct two-phase simulations of heterogeneous H₂turnover on the Pt(111) catalyst surface at chemical accuracy. The model trains itself in three days and performs at twice the speed of a ReaxFF model, while maintaining much higher fidelity to DFT and excellent agreement with experiment.