arXiv subjects

Pascal Friederich

Publications and source records attributed to Pascal Friederich.

At least 37 records · Page 2Linked to original sources

arXiv2024

PAL -- Parallel active learning for machine-learned potentials

Constructing datasets representative of the target domain is essential for training effective machine learning models. Active learning (AL) is a promising method that iteratively extends training data to enhance model performance while minimizing data acquisition costs. However, current AL workflows often require human intervention and lack parallelism, leading to inefficiencies and underutilization of modern computational resources. In this work, we introduce PAL, an automated, modular, and parallel active learning library that integrates AL tasks and manages their execution and communication on shared- and distributed-memory systems using the Message Passing Interface (MPI). PAL provides users with the flexibility to design and customize all components of their active learning scenarios, including machine learning models with uncertainty estimation, oracles for ground truth labeling, and strategies for exploring the target space. We demonstrate that PAL significantly reduces computational overhead and improves scalability, achieving substantial speed-ups through asynchronous parallelization on CPU and GPU hardware. Applications of PAL to several real-world scenarios - including ground-state reactions in biomolecular systems, excited-state dynamics of molecules, simulations of inorganic clusters, and thermo-fluid dynamics - illustrate its effectiveness in accelerating the development of machine learning models. Our results show that PAL enables efficient utilization of high-performance computing resources in active learning workflows, fostering advancements in scientific research and engineering applications.

cs.LG↗

arXiv2024

Discovering one molecule out of a million: inverse design of molecular hole transporting semiconductors tailored for perovskite solar cells

The inverse design of tailored organic molecules for specific optoelectronic devices of high complexity holds an enormous potential, but has not yet been realized1,2. The complexity and literally infinite diversity of conjugated molecular structures present both, an unprecedented opportunity for technological breakthroughs as well as an unseen optimization challenge. Current models rely on big data which do not exist for specialized research films. However, a hybrid computational and high throughput experimental screening workflow allowed us to train predictive models with as little as 149 molecules. We demonstrate a unique closed-loop workflow combining high throughput synthesis and Bayesian optimization that discovers new hole transporting materials with tailored properties for solar cell applications. A series of high-performance molecules were identified from minimal suggestions, achieving up to 26.23% (certified 25.88%) power conversion efficiency in perovskite solar cells. Our work paves the way for rapid, informed discovery in vast molecular libraries, revolutionizing material selection for complex devices. We believe that our approach can be generalized to other emerging fields and indeed accelerate the development of optoelectronic semiconductor devices in general.

cond-mat.mtrl-sci↗

arXiv2024

Conditional Normalizing Flows for Active Learning of Coarse-Grained Molecular Representations

Efficient sampling of the Boltzmann distribution of molecular systems is a long-standing challenge. Recently, instead of generating long molecular dynamics simulations, generative machine learning methods such as normalizing flows have been used to learn the Boltzmann distribution directly, without samples. However, this approach is susceptible to mode collapse and thus often does not explore the full configurational space. In this work, we address this challenge by separating the problem into two levels, the fine-grained and coarse-grained degrees of freedom. A normalizing flow conditioned on the coarse-grained space yields a probabilistic connection between the two levels. To explore the configurational space, we employ coarse-grained simulations with active learning which allows us to update the flow and make all-atom potential energy evaluations only when necessary. Using alanine dipeptide as an example, we show that our methods obtain a speedup to molecular dynamics simulations of approximately 15.9 to 216.2 compared to the speedup of 4.5 of the current state-of-the-art machine learning approach.

cs.LG↗

arXiv2024

Contextualized Policy Recovery: Modeling and Interpreting Medical Decisions with Adaptive Imitation Learning

Interpretable policy learning seeks to estimate intelligible decision policies from observed actions; however, existing models force a tradeoff between accuracy and interpretability, limiting data-driven interpretations of human decision-making processes. Fundamentally, existing approaches are burdened by this tradeoff because they represent the underlying decision process as a universal policy, when in fact human decisions are dynamic and can change drastically under different contexts. Thus, we develop Contextualized Policy Recovery (CPR), which re-frames the problem of modeling complex decision processes as a multi-task learning problem, where each context poses a unique task and complex decision policies can be constructed piece-wise from many simple context-specific policies. CPR models each context-specific policy as a linear map, and generates new policy models $\textit{on-demand}$ as contexts are updated with new observations. We provide two flavors of the CPR framework: one focusing on exact local interpretability, and one retaining full global interpretability. We assess CPR through studies on simulated and real data, achieving state-of-the-art performance on predicting antibiotic prescription in intensive care units ($+22\%$ AUROC vs. previous SOTA) and predicting MRI prescription for Alzheimer's patients ($+7.7\%$ AUROC vs. previous SOTA). With this improvement, CPR closes the accuracy gap between interpretable and black-box methods, allowing high-resolution exploration and analysis of context-specific decision models.

cs.LG↗

arXiv2024

Global Concept Explanations for Graphs by Contrastive Learning

Beyond improving trust and validating model fairness, xAI practices also have the potential to recover valuable scientific insights in application domains where little to no prior human intuition exists. To that end, we propose a method to extract global concept explanations from the predictions of graph neural networks to develop a deeper understanding of the tasks underlying structure-property relationships. We identify concept explanations as dense clusters in the self-explaining Megan models subgraph latent space. For each concept, we optimize a representative prototype graph and optionally use GPT-4 to provide hypotheses about why each structure has a certain effect on the prediction. We conduct computational experiments on synthetic and real-world graph property prediction tasks. For the synthetic tasks we find that our method correctly reproduces the structural rules by which they were created. For real-world molecular property regression and classification tasks, we find that our method rediscovers established rules of thumb. More specifically, our results for molecular mutagenicity prediction indicate more fine-grained resolution of structural details than existing explainability methods, consistent with previous results from chemistry literature. Overall, our results show promising capability to extract the underlying structure-property relationships for complex graph property prediction tasks.

cs.LG↗

arXiv2024

JARVIS-Leaderboard: A Large Scale Benchmark of Materials Design Methods

Lack of rigorous reproducibility and validation are major hurdles for scientific development across many fields. Materials science in particular encompasses a variety of experimental and theoretical approaches that require careful benchmarking. Leaderboard efforts have been developed previously to mitigate these issues. However, a comprehensive comparison and benchmarking on an integrated platform with multiple data modalities with both perfect and defect materials data is still lacking. This work introduces JARVIS-Leaderboard, an open-source and community-driven platform that facilitates benchmarking and enhances reproducibility. The platform allows users to set up benchmarks with custom tasks and enables contributions in the form of dataset, code, and meta-data submissions. We cover the following materials design categories: Artificial Intelligence (AI), Electronic Structure (ES), Force-fields (FF), Quantum Computation (QC) and Experiments (EXP). For AI, we cover several types of input data, including atomic structures, atomistic images, spectra, and text. For ES, we consider multiple ES approaches, software packages, pseudopotentials, materials, and properties, comparing results to experiment. For FF, we compare multiple approaches for material property predictions. For QC, we benchmark Hamiltonian simulations using various quantum algorithms and circuits. Finally, for experiments, we use the inter-laboratory approach to establish benchmarks. There are 1281 contributions to 274 benchmarks using 152 methods with more than 8 million data-points, and the leaderboard is continuously expanding. The JARVIS-Leaderboard is available at the website: https://pages.nist.gov/jarvis_leaderboard

cond-mat.mtrl-sci↗

arXiv2023

Neural networks trained on synthetically generated crystals can extract structural information from ICSD powder X-ray diffractograms

Machine learning techniques have successfully been used to extract structural information such as the crystal space group from powder X-ray diffractograms. However, training directly on simulated diffractograms from databases such as the ICSD is challenging due to its limited size, class-inhomogeneity, and bias toward certain structure types. We propose an alternative approach of generating synthetic crystals with random coordinates by using the symmetry operations of each space group. Based on this approach, we demonstrate online training of deep ResNet-like models on up to a few million unique on-the-fly generated synthetic diffractograms per hour. For our chosen task of space group classification, we achieved a test accuracy of 79.9% on unseen ICSD structure types from most space groups. This surpasses the 56.1% accuracy of the current state-of-the-art approach of training on ICSD crystals directly. Our results demonstrate that synthetically generated crystals can be used to extract structural information from ICSD powder diffractograms, which makes it possible to apply very large state-of-the-art machine learning models in the area of powder X-ray diffraction. We further show first steps toward applying our methodology to experimental data, where automated XRD data analysis is crucial, especially in high-throughput settings. While we focused on the prediction of the space group, our approach has the potential to be extended to related tasks in the future.

cond-mat.mtrl-sci↗

arXiv2023

Connectivity Optimized Nested Graph Networks for Crystal Structures

Graph neural networks (GNNs) have been applied to a large variety of applications in materials science and chemistry. Here, we recapitulate the graph construction for crystalline (periodic) materials and investigate its impact on the GNNs model performance. We suggest the asymmetric unit cell as a representation to reduce the number of atoms by using all symmetries of the system. This substantially reduced the computational cost and thus time needed to train large graph neural networks without any loss in accuracy. Furthermore, with a simple but systematically built GNN architecture based on message passing and line graph templates, we introduce a general architecture (Nested Graph Network, NGN) that is applicable to a wide range of tasks. We show that our suggested models systematically improve state-of-the-art results across all tasks within the MatBench benchmark. Further analysis shows that optimized connectivity and deeper message functions are responsible for the improvement. Asymmetric unit cells and connectivity optimization can be generally applied to (crystal) graph networks, while our suggested nested graph framework will open new ways of systematic comparison of GNN architectures.

cs.LG↗

arXiv2023

Machine learning for rapid discovery of laminar flow channel wall modifications that enhance heat transfer

Numerical simulation of fluids plays an essential role in modeling many physical phenomena, which enables technological advancements, contributes to sustainable practices, and expands our understanding of various natural and engineered systems. The calculation of heat transfer in fluid flow in simple flat channels is a relatively easy task for various simulation methods. However, once the channel geometry becomes more complex, numerical simulations become a bottleneck in optimizing wall geometries. We present a combination of accurate numerical simulations of arbitrary, flat, and non-flat channels and machine learning models predicting drag coefficient and Stanton number. We show that convolutional neural networks (CNN) can accurately predict the target properties at a fraction of the time of numerical simulations. We use the CNN models in a virtual high-throughput screening approach to explore a large number of possible, randomly generated wall architectures. Data Augmentation was applied to existing geometries data to add generated new training data which have the same number of parameters of heat transfer to improve the model's generalization. The general approach is not only applicable to simple flow setups as presented here but can be extended to more complex tasks, such as multiphase or even reactive unit operations in chemical engineering.

physics.flu-dyn↗

arXiv2023

MEGAN: Multi-Explanation Graph Attention Network

We propose a multi-explanation graph attention network (MEGAN). Unlike existing graph explainability methods, our network can produce node and edge attributional explanations along multiple channels, the number of which is independent of task specifications. This proves crucial to improve the interpretability of graph regression predictions, as explanations can be split into positive and negative evidence w.r.t to a reference value. Additionally, our attention-based network is fully differentiable and explanations can actively be trained in an explanation-supervised manner. We first validate our model on a synthetic graph regression dataset with known ground-truth explanations. Our network outperforms existing baseline explainability methods for the single- as well as the multi-explanation case, achieving near-perfect explanation accuracy during explanation supervision. Finally, we demonstrate our model's capabilities on multiple real-world datasets. We find that our model produces sparse high-fidelity explanations consistent with human intuition about those tasks.

cs.LG↗

arXiv2023

Quantifying the Intrinsic Usefulness of Attributional Explanations for Graph Neural Networks with Artificial Simulatability Studies

Despite the increasing relevance of explainable AI, assessing the quality of explanations remains a challenging issue. Due to the high costs associated with human-subject experiments, various proxy metrics are often used to approximately quantify explanation quality. Generally, one possible interpretation of the quality of an explanation is its inherent value for teaching a related concept to a student. In this work, we extend artificial simulatability studies to the domain of graph neural networks. Instead of costly human trials, we use explanation-supervisable graph neural networks to perform simulatability studies to quantify the inherent usefulness of attributional graph explanations. We perform an extensive ablation study to investigate the conditions under which the proposed analyses are most meaningful. We additionally validate our methods applicability on real-world graph classification and regression datasets. We find that relevant explanations can significantly boost the sample efficiency of graph neural networks and analyze the robustness towards noise and bias in the explanations. We believe that the notion of usefulness obtained from our proposed simulatability analysis provides a dimension of explanation quality that is largely orthogonal to the common practice of faithfulness and has great potential to expand the toolbox of explanation quality assessments, specifically for graph explanations.

cs.LG↗

arXiv2023

An integrated system built for small-molecule semiconductors via high-throughput approaches

High-throughput synthesis of solution-processable structurally variable small-molecule semiconductors is both an opportunity and a challenge. A large number of diverse molecules provide a possibility for quick material discovery and machine learning based on experimental data. However, the diversity of molecular structure leads to the complexity of molecular properties, such as solubility, polarity, and crystallinity, which poses great challenges to solution processing and purification. Here, we first report an integrated system for the high-throughput synthesis, purification, and characterization of molecules with a large variety. Based on the principle of Like dissolves like, we combine theoretical calculations and a robotic platform to accelerate the purification of those molecules. With this platform, a material library containing 125 molecules and their optical-electric properties was built within a timeframe of weeks. More importantly, the high repeatability of recrystallization we design is a reliable approach to further upgrading and industrial production.

physics.app-ph↗

arXiv2023

What is missing in autonomous discovery: Open challenges for the community

Self-driving labs (SDLs) leverage combinations of artificial intelligence, automation, and advanced computing to accelerate scientific discovery. The promise of this field has given rise to a rich community of passionate scientists, engineers, and social scientists, as evidenced by the development of the Acceleration Consortium and recent Accelerate Conference. Despite its strengths, this rapidly developing field presents numerous opportunities for growth, challenges to overcome, and potential risks of which to remain aware. This community perspective builds on a discourse instantiated during the first Accelerate Conference, and looks to the future of self-driving labs with a tempered optimism. Incorporating input from academia, government, and industry, we briefly describe the current status of self-driving labs, then turn our attention to barriers, opportunities, and a vision for what is possible. Our field is delivering solutions in technology and infrastructure, artificial intelligence and knowledge generation, and education and workforce development. In the spirit of community, we intend for this work to foster discussion and drive best practices as our field grows.

cond-mat.mtrl-sci↗

arXiv2023

Accurate GW frontier orbital energies of 134 kilo molecules

The QM9 dataset [Scientific Data, Vol. 1, 140022 (2014)] became a standard dataset to benchmark machine learning methods, especially on molecular graphs. It contains geometries as well as multiple computed molecular properties of 133,885 compounds at B3LYP/6-31G(2df,p) level of theory, including frontier orbitals (HOMO and LUMO) energies. However, the accuracy of HOMO/LUMO predictions from density functional theory, including hybrid methods such as B3LYP, is limited for many applications. In contrast, the GW method significantly improves HOMO/LUMO prediction accuracy, with mean unsigned errors in the GW100 benchmark dataset of 100 meV. In this work, we present a new dataset of HOMO/LUMO energies for the QM9 compounds, computed using the GW method. This database may serve as a benchmark of HOMO/LUMO prediction, delta-learning, and transfer learning, particularly for larger molecules where GW is the most accurate but still numerically feasible method. We expect this dataset to enable the development of more accurate machine learning models for predicting molecular properties

physics.chem-ph↗

arXiv2022

3DSC - A New Dataset of Superconductors Including Crystal Structures

Data-driven methods, in particular machine learning, can help to speed up the discovery of new materials by finding hidden patterns in existing data and using them to identify promising candidate materials. In the case of superconductors, which are a highly interesting but also a complex class of materials with many relevant applications, the use of data science tools is to date slowed down by a lack of accessible data. In this work, we present a new and publicly available superconductivity dataset ('3DSC'), featuring the critical temperature $T_\mathrm{c}$ of superconducting materials additionally to tested non-superconductors. In contrast to existing databases such as the SuperCon database which contains information on the chemical composition, the 3DSC is augmented by the approximate three-dimensional crystal structure of each material. We perform a statistical analysis and machine learning experiments to show that access to this structural information improves the prediction of the critical temperature $T_\mathrm{c}$ of materials. Furthermore, we see the 3DSC not as a finished dataset, but we provide ideas and directions for further research to improve the 3DSC in multiple ways. We are confident that this database will be useful in applying state-of-the-art machine learning methods to eventually find new superconductors.

cond-mat.supr-con↗

arXiv2022

Actively Learning Costly Reward Functions for Reinforcement Learning

Transfer of recent advances in deep reinforcement learning to real-world applications is hindered by high data demands and thus low efficiency and scalability. Through independent improvements of components such as replay buffers or more stable learning algorithms, and through massively distributed systems, training time could be reduced from several days to several hours for standard benchmark tasks. However, while rewards in simulated environments are well-defined and easy to compute, reward evaluation becomes the bottleneck in many real-world environments, e.g., in molecular optimization tasks, where computationally demanding simulations or even experiments are required to evaluate states and to quantify rewards. Therefore, training might become prohibitively expensive without an extensive amount of computational resources and time. We propose to alleviate this problem by replacing costly ground-truth rewards with rewards modeled by neural networks, counteracting non-stationarity of state and reward distributions during training with an active learning component. We demonstrate that using our proposed ACRL method (Actively learning Costly rewards for Reinforcement Learning), it is possible to train agents in complex real-world environments orders of magnitudes faster. By enabling the application of reinforcement learning methods to new domains, we show that we can find interesting and non-trivial solutions to real-world optimization problems in chemistry, materials science and engineering.

cs.LG↗

arXiv2022

Graph neural networks to learn joint representations of disjoint molecular graphs

Graph neural networks are widely used to learn global representations of graphs, which are then used for regression or classification tasks. Typically, the graphs in such data sets are connected, i.e. each training sample consists of a single internally connected graph associated with a global label. However, there is a wide variety of yet unconsidered but application-relevant tasks, where labels are assigned to sets of disjoint graphs, which requires the generation of global representations of disjoint graphs. In this paper, we present a new data set with chemical reactions, which is illustrating this task. Each sample consists of a pair of disjoint molecular graphs and a joint label representing a scalar measure associated with the chemical reaction of the molecules. We show the initial results of graph neural networks that are able to solve the task within a combinatorial subset of the dataset but do not generalize well to the full data set and unseen (sub)graphs.

cs.CE↗

arXiv2022

Graph neural networks for materials science and chemistry

Machine learning plays an increasingly important role in many areas of chemistry and materials science, e.g. to predict materials properties, to accelerate simulations, to design new materials, and to predict synthesis routes of new materials. Graph neural networks (GNNs) are one of the fastest growing classes of machine learning models. They are of particular relevance for chemistry and materials science, as they directly work on a graph or structural representation of molecules and materials and therefore have full access to all relevant information required to characterize materials. In this review article, we provide an overview of the basic principles of GNNs, widely used datasets, and state-of-the-art architectures, followed by a discussion of a wide range of recent applications of GNNs in chemistry and materials science, and concluding with a road-map for the further development and application of GNNs.

physics.chem-ph↗