In-Context Learning of Physical Properties: Few-Shot Adaptation to Out-of-Distribution Molecular Graphs

Large language models manifest the ability of few-shot adaptation to a sequence of provided examples. This behavior, known as in-context learning, allows for performing nontrivial machine learning tasks during inference only. In this work, we address the question: can we leverage in-context learning to predict out-of-distribution materials properties? However, this would not be possible for structure property prediction tasks unless an effective method is found to pass atomic-level geometric features to the transformer model. To address this problem, we employ a compound model in which GPT-2 acts on the output of geometry-aware graph neural networks to adapt in-context information. To demonstrate our model's capabilities, we partition the QM9 dataset into sequences of molecules that share a common substructure and use them for in-context learning. This approach significantly improves the performance of the model on out-of-distribution examples, surpassing the one of general graph neural network models.

Substitutional Alloying Using Crystal Graph Neural Networks

Materials discovery, especially for applications that require extreme operating conditions, requires extensive testing that naturally limits the ability to inquire the wealth of possible compositions. Machine Learning (ML) has nowadays a well established role in facilitating this effort in systematic ways. The increasing amount of available accurate DFT data represents a solid basis upon which new ML models can be trained and tested. While conventional models rely on static descriptors, generally suitable for a limited class of systems, the flexibility of Graph Neural Networks (GNNs) allows for direct learning representations on graphs, such as the ones formed by crystals. We utilize crystal graph neural networks (CGNN) to predict crystal properties with DFT level accuracy, through graphs with encoding of the atomic (node/vertex), bond (edge), and global state attributes. In this work, we aim at testing the ability of the CGNN MegNet framework in predicting a number of properties of systems previously unseen from the model, obtained by adding a substitutional defect in bulk crystals that are included in the training set. We perform DFT validation to assess the accuracy in the prediction of formation energies and structural features (such as elastic moduli). Using CGNNs, one may identify promising paths in alloy discovery.

Direct detection of plasticity onset through total-strain profile evolution

Plastic yielding in solids strongly depends on various conditions, such as temperature and loading rate and indeed, sample-dependent knowledge of yield points in structural materials promotes reliability in mechanical behavior. Commonly, yielding is measured through controlled mechanical testing at small or large scales, in ways that either distinguish elastic (stress) from total deformation measurements, or by identifying plastic slip contributions. In this paper we argue that instead of separate elastic/plastic measurements, yielding can be unraveled through statistical analysis of total strain fluctuations during the evolution sequence of profiles measured in-situ, through digital image correlation. We demonstrate two distinct ways of precisely quantifying yield locations in widely applicable crystal plasticity models, that apply in polycrystalline solids, either by using principal component analysis or discrete wavelet transforms. We test and compare these approaches in synthetic data of polycrystal simulations and a variety of yielding responses, through changes of the applied loading rates and the strain-rate sensitivity exponents.

Learning crystal plasticity using digital image correlation: Examples from discrete dislocation dynamics

Digital image correlation (DIC) is a well-established, non-invasive technique for tracking and quantifying the deformation of mechanical samples under strain. While it provides an obvious way to observe incremental and aggregate displacement information, it seems likely that DIC data sets, which after all reflect the spatially-resolved response of a microstructure to loads, contain much richer information than has generally been extracted from them. In this paper, we demonstrate a machine-learning approach to quantifying the prior deformation history of a crystalline sample based on its response to a subsequent DIC test. This prior deformation history is encoded in the microstructure through the inhomogeneity of the dislocation microstructure, and in the spatial correlations of the dislocation patterns, which mediate the system's response to the DIC test load. Our domain consists of deformed crystalline thin films generated by a discrete dislocation plasticity simulation. We explore the range of applicability of machine learning (ML) for typical experimental protocols, and as a function of possible size effects and stochasticity. Plasticity size effects may directly influence the data, rendering unsupervised techniques unable to distinguish different plasticity regimes.