{"topic":"interpretability","sort":"top","limit":25,"offset":0,"count":25,"papers":[{"id":"doi:10.1038/nrg3934","kind":"review","title":"Systems biology and gene networks in neurodevelopmental and neurodegenerative disorders","authors":["Neelroop Parikshak","Michael J. Gandal","Daniel H. Geschwind"],"author_count":3,"year":2015,"published":"2015-07-07","venue":"Nature Reviews Genetics","cited_by":447,"doi":"10.1038/nrg3934","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1038/nrg3934","archive_url":"https://space-kitty.com/papers/doi:10.1038/nrg3934","updated":null,"pmid":"26149713","pmcid":null,"openalex_id":"W1508717233","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":false,"peer_reviewed":null,"categories":["Bioinformatics and Genomic Networks"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"doi:10.1038/s41592-019-0576-7","kind":"article","title":"Exploring single-cell data with deep multitasking neural networks","authors":["Matthew Amodio","David van Dijk","Krishnan Srinivasan","William S. Chen","Hussein Mohsen","Kevin R. Moon","Allison M. Campbell","Yujiao Zhao","Xiaomei Wang","Manjunatha M. Venkataswamy","Anita Sudhir Desai","Vasanthapuram Ravi","Priti Kumar","Ruth R. Montgomery","Guy Wolf","Smita Krishnaswamy"],"author_count":16,"year":2019,"published":"2019-10-07","venue":"Nature Methods","cited_by":297,"doi":"10.1038/s41592-019-0576-7","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1038/s41592-019-0576-7","archive_url":"https://space-kitty.com/papers/doi:10.1038/s41592-019-0576-7","updated":null,"pmid":"31591579","pmcid":null,"openalex_id":"W2978541146","pdf_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC10164410/pdf/nihms-1891783.pdf","abstract":null,"abstract_license":null,"license":null,"is_open_access":true,"peer_reviewed":null,"categories":["Single-cell and spatial transcriptomics"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"doi:10.1126/science.aam5887","kind":"article","title":"Decoding of position in the developing neural tube from antiparallel morphogen gradients","authors":["Marcin Zagórski","Yoji Tabata","Nathalie Brandenberg","Matthias P. Lütolf","Gašper Tkačik","Tobias Bollenbach","James Briscoe","Anna Kicheva"],"author_count":8,"year":2017,"published":"2017-06-30","venue":"Science","cited_by":184,"doi":"10.1126/science.aam5887","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1126/science.aam5887","archive_url":"https://space-kitty.com/papers/doi:10.1126/science.aam5887","updated":null,"pmid":"28663499","pmcid":null,"openalex_id":"W2725924882","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":true,"peer_reviewed":null,"categories":["Developmental Biology and Gene Regulation"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"arxiv:2005.13073","kind":"article","title":"Biologically-informed neural networks guide mechanistic modeling from sparse experimental data","authors":["John H. Lagergren","John T. Nardini","Ruth E. Baker","Matthew J. Simpson","Kevin Flores"],"author_count":5,"year":2020,"published":"2020-12-01","venue":"PLoS Computational Biology","cited_by":167,"doi":"10.1371/journal.pcbi.1008462","arxiv_id":"2005.13073","nct_id":null,"source_url":"https://doi.org/10.1371/journal.pcbi.1008462","archive_url":"https://space-kitty.com/papers/arxiv:2005.13073","updated":"2020-05-26","pmid":"33259472","pmcid":null,"openalex_id":"W3029447859","pdf_url":"https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1008462&type=printable","abstract":"Biologically-informed neural networks (BINNs), an extension of physics-informed neural networks [1], are introduced and used to discover the underlying dynamics of biological systems from sparse experimental data. In the present work, BINNs are trained in a supervised learning framework to approximate in vitro cell biology assay experiments while respecting a generalized form of the governing reaction-diffusion partial differential equation (PDE). By allowing the diffusion and reaction terms to be multilayer perceptrons (MLPs), the nonlinear forms of these terms can be learned while simultaneously converging to the solution of the governing PDE. Further, the trained MLPs are used to guide the selection of biologically interpretable mechanistic forms of the PDE terms which provides new insights into the biological and physical mechanisms that govern the dynamics of the observed system. The method is evaluated on sparse real-world data from wound healing assays with varying initial cell densities [2].","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Model Reduction and Neural Networks"],"details":null,"provenance":"openalex+arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv."},{"id":"doi:10.1186/s13059-020-02100-5","kind":"article","title":"Knowledge-primed neural networks enable biologically interpretable deep learning on single-cell sequencing data","authors":["Nikolaus Fortelny","Christoph Bock"],"author_count":2,"year":2020,"published":"2020-08-03","venue":"Genome biology","cited_by":135,"doi":"10.1186/s13059-020-02100-5","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1186/s13059-020-02100-5","archive_url":"https://space-kitty.com/papers/doi:10.1186/s13059-020-02100-5","updated":null,"pmid":"32746932","pmcid":null,"openalex_id":"W3047163763","pdf_url":"https://genomebiology.biomedcentral.com/track/pdf/10.1186/s13059-020-02100-5","abstract":"BACKGROUND: Deep learning has emerged as a versatile approach for predicting complex biological phenomena. However, its utility for biological discovery has so far been limited, given that generic deep neural networks provide little insight into the biological mechanisms that underlie a successful prediction. Here we demonstrate deep learning on biological networks, where every node has a molecular equivalent, such as a protein or gene, and every edge has a mechanistic interpretation, such as a regulatory interaction along a signaling pathway. RESULTS: With knowledge-primed neural networks (KPNNs), we exploit the ability of deep learning algorithms to assign meaningful weights in multi-layered networks, resulting in a widely applicable approach for interpretable deep learning. We present a learning method that enhances the interpretability of trained KPNNs by stabilizing node weights in the presence of redundancy, enhancing the quantitative interpretability of node weights, and controlling for uneven connectivity in biological networks. We validate KPNNs on simulated data with known ground truth and demonstrate their practical use and utility in five biological applications with single-cell RNA-seq data for cancer and immune cells. CONCLUSIONS: We introduce KPNNs as a method that combines the predictive power of deep learning with the interpretability of biological networks. While demonstrated here on single-cell sequencing data, this method is broadly relevant to other research areas where prior domain knowledge can be represented as networks.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Bioinformatics and Genomic Networks"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Knowledge-primed neural networks enable biologically interpretable deep learning on single-cell sequencing data\" by Nikolaus Fortelny, Christoph Bock, https://doi.org/10.1186/s13059-020-02100-5, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.1016/j.watres.2023.120667","kind":"article","title":"Machine learning for modeling N2O emissions from wastewater treatment plants: Aligning model performance, complexity, and interpretability","authors":["Mostafa Khalil","Ahmed AlSayed","Yang Liu","Peter A. Vanrolleghem"],"author_count":4,"year":2023,"published":"2023-09-24","venue":"Water Research","cited_by":116,"doi":"10.1016/j.watres.2023.120667","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1016/j.watres.2023.120667","archive_url":"https://space-kitty.com/papers/doi:10.1016/j.watres.2023.120667","updated":null,"pmid":"37778084","pmcid":null,"openalex_id":"W4386999101","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":false,"peer_reviewed":null,"categories":["Wastewater Treatment and Nitrogen Removal"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"doi:10.5194/hess-26-5085-2022","kind":"article","title":"Improving hydrologic models for predictions and process understanding using neural ODEs","authors":["Marvin Höge","Andreas Scheidegger","Marco Baity‐Jesi","Carlo Albert","Fabrizio Fenicia"],"author_count":5,"year":2022,"published":"2022-10-11","venue":"Hydrology and earth system sciences","cited_by":126,"doi":"10.5194/hess-26-5085-2022","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.5194/hess-26-5085-2022","archive_url":"https://space-kitty.com/papers/doi:10.5194/hess-26-5085-2022","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W4304191791","pdf_url":"https://hess.copernicus.org/articles/26/5085/2022/hess-26-5085-2022.pdf","abstract":"Abstract. Deep learning methods have frequently outperformed conceptual hydrologic models in rainfall-runoff modelling. Attempts of investigating the internals of such deep learning models are being made but traceability of model states and processes and their interrelations to model input and output is not yet fully given. Direct interpretability of mechanistic processes has always been considered as asset of conceptual models that helps to gain system understanding aside of predictability. We introduce hydrologic Neural Ordinary Differential Equation (ODE) models that perform as well as state-of-the-art deep learning methods in stream flow prediction while maintaining the ease of interpretability of conceptual hydrologic models. In Neural ODEs, internal processes that are represented in differential equations are substituted by neural networks. Therefore, Neural ODE models enable fusing deep learning with mechanistic modelling. We demonstrate the basin-specific predictive performance for several hundred catchments of the continental USA. For exemplary basins, we analyse the dynamics of states and processes learned by the model-internal neural networks. Finally, we discuss the potential of Neural ODE models in hydrology.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Hydrological Forecasting Using AI"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Improving hydrologic models for predictions and process understanding using neural ODEs\" by Marvin Höge, Andreas Scheidegger, Marco Baity‐Jesi et al., https://doi.org/10.5194/hess-26-5085-2022, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.1016/j.conbuildmat.2018.09.087","kind":"article","title":"Improving accuracy of rutting prediction for mechanistic-empirical pavement design guide with deep neural networks","authors":["Hongren Gong","Yiren Sun","Zijun Mei","Baoshan Huang"],"author_count":4,"year":2018,"published":"2018-09-28","venue":"Construction and Building Materials","cited_by":130,"doi":"10.1016/j.conbuildmat.2018.09.087","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1016/j.conbuildmat.2018.09.087","archive_url":"https://space-kitty.com/papers/doi:10.1016/j.conbuildmat.2018.09.087","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W2893772597","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":false,"peer_reviewed":null,"categories":["Asphalt Pavement Performance Evaluation"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"arxiv:2310.19998","kind":"article","title":"Generative Retrieval-Augmented Ontologic Graph and Multiagent Strategies for Interpretive Large Language Model-Based Materials Design","authors":["Markus J. Buehler"],"author_count":1,"year":2024,"published":"2024-01-12","venue":"ACS Engineering Au","cited_by":82,"doi":"10.1021/acsengineeringau.3c00058","arxiv_id":"2310.19998","nct_id":null,"source_url":"https://doi.org/10.1021/acsengineeringau.3c00058","archive_url":"https://space-kitty.com/papers/arxiv:2310.19998","updated":"2023-10-30","pmid":"38646516","pmcid":null,"openalex_id":"W4390824394","pdf_url":"https://arxiv.org/pdf/2310.19998","abstract":"Transformer neural networks show promising capabilities, in particular for uses in materials analysis, design and manufacturing, including their capacity to work effectively with both human language, symbols, code, and numerical data. Here we explore the use of large language models (LLMs) as a tool that can support engineering analysis of materials, applied to retrieving key information about subject areas, developing research hypotheses, discovery of mechanistic relationships across disparate areas of knowledge, and writing and executing simulation codes for active knowledge generation based on physical ground truths. When used as sets of AI agents with specific features, capabilities, and instructions, LLMs can provide powerful problem solution strategies for applications in analysis and design problems. Our experiments focus on using a fine-tuned model, MechGPT, developed based on training data in the mechanics of materials domain. We first affirm how finetuning endows LLMs with reasonable understanding of domain knowledge. However, when queried outside the context of learned matter, LLMs can have difficulty to recall correct information. We show how this can be addressed using retrieval-augmented Ontological Knowledge Graph strategies that discern how the model understands what concepts are important and how they are related. Illustrated for a use case of relating distinct areas of knowledge - here, music and proteins - such strategies can also provide an interpretable graph structure with rich information at the node, edge and subgraph level. We discuss nonlinear sampling strategies and agent-based modeling applied to complex question answering, code generation and execution in the context of automated force field development from actively learned Density Functional Theory (DFT) modeling, and data analysis.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":"cc-by-nc-nd","is_open_access":true,"peer_reviewed":null,"categories":["Machine Learning in Materials Science"],"details":null,"provenance":"openalex+arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv."},{"id":"doi:10.1016/j.devcel.2012.09.015","kind":"article","title":"Mechanistic Differences in the Transcriptional Interpretation of Local and Long-Range Shh Morphogen Signaling","authors":["Tony Oosterveen","Sanja Kurdija","Zhanna Alekseenko","Christopher W. Uhde","Maria Bergsland","Magnus Sandberg","Elisabet Andersson","José Maria Moreira Dias","Jonas Muhr","Johan Ericson"],"author_count":10,"year":2012,"published":"2012-11-01","venue":"Developmental Cell","cited_by":132,"doi":"10.1016/j.devcel.2012.09.015","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1016/j.devcel.2012.09.015","archive_url":"https://space-kitty.com/papers/doi:10.1016/j.devcel.2012.09.015","updated":null,"pmid":"23153497","pmcid":null,"openalex_id":"W1978292443","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":true,"peer_reviewed":null,"categories":["Hedgehog Signaling Pathway Studies"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"doi:10.1371/journal.pcbi.1008472","kind":"article","title":"Development of a hybrid model for a partially known intracellular signaling pathway through correction term estimation and neural network modeling","authors":["Dongheon Lee","Arul Jayaraman","Joseph Sang‐Il Kwon"],"author_count":3,"year":2020,"published":"2020-12-14","venue":"PLoS Computational Biology","cited_by":103,"doi":"10.1371/journal.pcbi.1008472","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1371/journal.pcbi.1008472","archive_url":"https://space-kitty.com/papers/doi:10.1371/journal.pcbi.1008472","updated":null,"pmid":"33315899","pmcid":null,"openalex_id":"W3113010729","pdf_url":"https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1008472&type=printable","abstract":"Developing an accurate first-principle model is an important step in employing systems biology approaches to analyze an intracellular signaling pathway. However, an accurate first-principle model is difficult to be developed since it requires in-depth mechanistic understandings of the signaling pathway. Since underlying mechanisms such as the reaction network structure are not fully understood, significant discrepancy exists between predicted and actual signaling dynamics. Motivated by these considerations, this work proposes a hybrid modeling approach that combines a first-principle model and an artificial neural network (ANN) model so that predictions of the hybrid model surpass those of the original model. First, the proposed approach determines an optimal subset of model states whose dynamics should be corrected by the ANN by examining the correlation between each state and outputs through relative order. Second, an L2-regularized least-squares problem is solved to infer values of the correction terms that are necessary to minimize the discrepancy between the model predictions and available measurements. Third, an ANN is developed to generalize relationships between the values of the correction terms and the system dynamics. Lastly, the original first-principle model is coupled with the developed ANN to finalize the hybrid model development so that the model will possess generalized prediction capabilities while retaining the model interpretability. We have successfully validated the proposed methodology with two case studies, simplified apoptosis and lipopolysaccharide-induced NFκB signaling pathways, to develop hybrid models with in silico and in vitro measurements, respectively.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Gene Regulatory Network Analysis"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Development of a hybrid model for a partially known intracellular signaling pathway through correction term estimation and neural network modeling\" by Dongheon Lee, Arul Jayaraman, Joseph Sang‐Il Kwon, https://doi.org/10.1371/journal.pcbi.1008472, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.3389/fimmu.2024.1363144","kind":"review","title":"A review of mechanistic learning in mathematical oncology","authors":["John Metzcar","Catherine Ruth Jutzeler","Paul Macklin","Alvaro Köhn‐Luque","Sarah C. Brüningk"],"author_count":5,"year":2024,"published":"2024-03-12","venue":"Frontiers in Immunology","cited_by":70,"doi":"10.3389/fimmu.2024.1363144","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.3389/fimmu.2024.1363144","archive_url":"https://space-kitty.com/papers/doi:10.3389/fimmu.2024.1363144","updated":null,"pmid":"38533513","pmcid":null,"openalex_id":"W4392714280","pdf_url":"https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2024.1363144/pdf","abstract":"Mechanistic learning refers to the synergistic combination of mechanistic mathematical modeling and data-driven machine or deep learning. This emerging field finds increasing applications in (mathematical) oncology. This review aims to capture the current state of the field and provides a perspective on how mechanistic learning may progress in the oncology domain. We highlight the synergistic potential of mechanistic learning and point out similarities and differences between purely data-driven and mechanistic approaches concerning model complexity, data requirements, outputs generated, and interpretability of the algorithms and their results. Four categories of mechanistic learning (sequential, parallel, extrinsic, intrinsic) of mechanistic learning are presented with specific examples. We discuss a range of techniques including physics-informed neural networks, surrogate model learning, and digital twins. Example applications address complex problems predominantly from the domain of oncology research such as longitudinal tumor response predictions or time-to-event modeling. As the field of mechanistic learning advances, we aim for this review and proposed categorization framework to foster additional collaboration between the data- and knowledge-driven modeling fields. Further collaboration will help address difficult issues in oncology such as limited data availability, requirements of model transparency, and complex input data which are embraced in a mechanistic learning framework.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Mathematical Biology Tumor Growth"],"details":null,"provenance":"openalex","attribution":"Abstract of \"A review of mechanistic learning in mathematical oncology\" by John Metzcar, Catherine Ruth Jutzeler, Paul Macklin et al., https://doi.org/10.3389/fimmu.2024.1363144, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.1126/sciadv.aax5979","kind":"article","title":"Efficient inverse graphics in biological face processing","authors":["Ilker Yildirim","Mario Belledonne","Winrich A. Freiwald","Joshua B. Tenenbaum"],"author_count":4,"year":2020,"published":"2020-03-04","venue":"Science Advances","cited_by":89,"doi":"10.1126/sciadv.aax5979","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1126/sciadv.aax5979","archive_url":"https://space-kitty.com/papers/doi:10.1126/sciadv.aax5979","updated":null,"pmid":"32181338","pmcid":null,"openalex_id":"W3010134765","pdf_url":"https://www.science.org/doi/pdf/10.1126/sciadv.aax5979?download=true","abstract":null,"abstract_license":null,"license":"cc-by-nc","is_open_access":true,"peer_reviewed":null,"categories":["Face Recognition and Perception"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"doi:10.3389/fnins.2022.906290","kind":"review","title":"Explainable AI: A review of applications to neuroimaging data","authors":["Farzad V. Farahani","Krzysztof Fiok","Behshad Lahijanian","Waldemar Karwowski","Pamela K. Douglas"],"author_count":5,"year":2022,"published":"2022-12-01","venue":"Frontiers in Neuroscience","cited_by":81,"doi":"10.3389/fnins.2022.906290","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.3389/fnins.2022.906290","archive_url":"https://space-kitty.com/papers/doi:10.3389/fnins.2022.906290","updated":null,"pmid":"36583102","pmcid":null,"openalex_id":"W4310525312","pdf_url":"https://www.frontiersin.org/articles/10.3389/fnins.2022.906290/pdf","abstract":"Deep neural networks (DNNs) have transformed the field of computer vision and currently constitute some of the best models for representations learned via hierarchical processing in the human brain. In medical imaging, these models have shown human-level performance and even higher in the early diagnosis of a wide range of diseases. However, the goal is often not only to accurately predict group membership or diagnose but also to provide explanations that support the model decision in a context that a human can readily interpret. The limited transparency has hindered the adoption of DNN algorithms across many domains. Numerous explainable artificial intelligence (XAI) techniques have been developed to peer inside the “black box” and make sense of DNN models, taking somewhat divergent approaches. Here, we suggest that these methods may be considered in light of the interpretation goal, including functional or mechanistic interpretations, developing archetypal class instances, or assessing the relevance of certain features or mappings on a trained model in a post-hoc capacity. We then focus on reviewing recent applications of post-hoc relevance techniques as applied to neuroimaging data. Moreover, this article suggests a method for comparing the reliability of XAI methods, especially in deep neural networks, along with their advantages and pitfalls.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Explainable Artificial Intelligence (XAI)"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Explainable AI: A review of applications to neuroimaging data\" by Farzad V. Farahani, Krzysztof Fiok, Behshad Lahijanian et al., https://doi.org/10.3389/fnins.2022.906290, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.14573/altex.2309191","kind":"article","title":"ToxAIcology - The evolving role of artificial intelligence in advancing toxicology and modernizing regulatory science","authors":["Thomas Härtung"],"author_count":1,"year":2023,"published":"2023-01-01","venue":"ALTEX","cited_by":77,"doi":"10.14573/altex.2309191","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.14573/altex.2309191","archive_url":"https://space-kitty.com/papers/doi:10.14573/altex.2309191","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W4388188339","pdf_url":"https://www.altex.org/index.php/altex/article/download/2692/2593","abstract":"Toxicology has undergone a transformation from an observational science to a data-rich discipline ripe for artificial intelligence (AI) integration. The exponential growth in computing power coupled with accumulation of large toxicological datasets has created new opportunities to apply techniques like machine learning and especially deep learning to enhance chemical hazard assessment. This article provides an overview of key developments in AI-enabled toxicology, including early expert systems, statistical learning methods like quantitative structure-activity relationships (QSARs), recent advances with deep neural networks, and emerging trends. The promises and challenges of AI adoption for predictive toxicology, data analysis, risk assessment, and mechanistic research are discussed. Responsible development and application of interpretable and human-centered AI tools through multidisciplinary collaboration can accelerate evidence-based toxicology to better protect human health and the environment. However, AI is not a panacea and must be thoughtfully designed and utilized alongside ongoing efforts to improve primary evidence generation and appraisal.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Computational Drug Discovery Methods"],"details":null,"provenance":"openalex","attribution":"Abstract of \"ToxAIcology - The evolving role of artificial intelligence in advancing toxicology and modernizing regulatory science\" by Thomas Härtung, https://doi.org/10.14573/altex.2309191, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.3390/ijms26199384","kind":"review","title":"AI-Integrated QSAR Modeling for Enhanced Drug Discovery: From Classical Approaches to Deep Learning and Structural Insight","authors":["Mahesh Koirala","Lindy Yan","Zoser Mohamed","Mario DiPaola"],"author_count":4,"year":2025,"published":"2025-09-25","venue":"International Journal of Molecular Sciences","cited_by":58,"doi":"10.3390/ijms26199384","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.3390/ijms26199384","archive_url":"https://space-kitty.com/papers/doi:10.3390/ijms26199384","updated":null,"pmid":"41096653","pmcid":null,"openalex_id":"W4414547037","pdf_url":null,"abstract":"Integrating artificial intelligence (AI) with the Quantitative Structure-Activity Relationship (QSAR) has transformed modern drug discovery by empowering faster, more accurate, and scalable identification of therapeutic compounds. This review outlines the evolution from classical QSAR methods, such as multiple linear regression and partial least squares, to advanced machine learning and deep learning approaches, including graph neural networks and SMILES-based transformers. Molecular docking and molecular dynamics simulations are presented as cooperative tools that boost the mechanistic consideration and structural insight into the ligand-target interactions. Discussions on using PROTACs and targeted protein degradation, ADMET prediction, and public databases and cloud-based platforms to democratize access to computational modeling are well presented with priority. Challenges related to authentication, interpretability, regulatory standards, and ethical concerns are examined, along with emerging patterns in AI-driven drug development. This review is a guideline for using computational models and databases in explainable, data-rich and profound drug discovery pipelines.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Computational Drug Discovery Methods"],"details":null,"provenance":"openalex","attribution":"Abstract of \"AI-Integrated QSAR Modeling for Enhanced Drug Discovery: From Classical Approaches to Deep Learning and Structural Insight\" by Mahesh Koirala, Lindy Yan, Zoser Mohamed et al., https://doi.org/10.3390/ijms26199384, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"arxiv:2304.14997","kind":"article","title":"Towards Automated Circuit Discovery for Mechanistic Interpretability","authors":["Arthur Conmy","Augustine N. Mavor-Parker","Aengus Lynch","Stefan Heimersheim","Adrià Garriga-Alonso"],"author_count":5,"year":2023,"published":"2023-04-28","venue":"neural information processing systems","cited_by":72,"doi":"10.52202/075280-0719","arxiv_id":"2304.14997","nct_id":null,"source_url":"https://arxiv.org/abs/2304.14997","archive_url":"https://space-kitty.com/papers/arxiv:2304.14997","updated":"2023-10-28","pmid":null,"pmcid":null,"openalex_id":"W7133218086","pdf_url":"https://arxiv.org/pdf/2304.14997","abstract":"Through considerable effort and intuition, several recent works have reverse-engineered nontrivial behaviors of transformer models. This paper systematizes the mechanistic interpretability process they followed. First, researchers choose a metric and dataset that elicit the desired model behavior. Then, they apply activation patching to find which abstract neural network units are involved in the behavior. By varying the dataset, metric, and units under investigation, researchers can understand the functionality of each component. We automate one of the process' steps: to identify the circuit that implements the specified behavior in the model's computational graph. We propose several algorithms and reproduce previous interpretability results to validate them. For example, the ACDC algorithm rediscovered 5/5 of the component types in a circuit in GPT-2 Small that computes the Greater-Than operation. ACDC selected 68 of the 32,000 edges in GPT-2 Small, all of which were manually found by previous work. Our code is available at https://github.com/ArthurConmy/Automatic-Circuit-Discovery.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":null,"is_open_access":false,"peer_reviewed":null,"categories":["cs.LG"],"details":null,"provenance":"arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv."},{"id":"doi:10.1016/j.cosrev.2025.100883","kind":"article","title":"A survey of transformer networks for time series forecasting","authors":["Jingyuan Zhao","Fulin Chu","Lili Xie","Yunhong Che","Yuyan Wu","Andrew F. Burke"],"author_count":6,"year":2025,"published":"2025-12-19","venue":"Computer Science Review","cited_by":54,"doi":"10.1016/j.cosrev.2025.100883","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1016/j.cosrev.2025.100883","archive_url":"https://space-kitty.com/papers/doi:10.1016/j.cosrev.2025.100883","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W4417506931","pdf_url":null,"abstract":"• Introduction to Transformers, their framework, hyperparameters, and learning paradigms. • Comprehensive assessment of tailored Transformers for advanced time series forecasting. • In-depth analysis of applications in finance, healthcare, energy, and transportation. • Discussion of challenges such as computational cost, datasets, overfitting, and interpretability. • Insights into integrating knowledge, multimodal, efficiency and AI-driven approaches. Time-series data support critical functions in forecasting, anomaly detection, resource optimization, and real-time decision making across finance, healthcare, energy systems, and networked computing. Classical statistical approaches and early deep-learning architectures (RNNs, LSTMs, CNNs) have achieved notable progress, yet they exhibit structural limitations: recurrent models struggle with long-range dependencies, convolutional models require deep stacking to enlarge receptive fields, and both scale suboptimally to high-dimensional and high-volume data. Transformer architectures—characterized by global attention mechanisms, flexible temporal receptive fields, and growing adoption within foundation-model paradigms—have consequently gained increasing prominence for modern time-series analysis. Drawing on a systematic review of IEEE Xplore, ACM Digital Library, and Scopus (2020–2025), this survey offers a unified, theoretically grounded synthesis of Transformer-based methods for time-series learning. The survey summarizes defining characteristics of time-series data and analyzes core architectural elements, including attention formulations, encoder–decoder structures, hyperparameter design, and domain-specific adaptations. An architecture-centered and task-aware taxonomy is presented to organize recent advances across forecasting, representation learning, anomaly detection, and multimodal fusion. Persistent challenges—spanning computational scalability, data-efficiency constraints, distributional heterogeneity, overfitting risks, hyperparameter instability, interpretability, and reproducibility—are examined in depth. A forward-looking research agenda is outlined, highlighting opportunities in physics-informed architectural design, hybrid neural–mechanistic modeling, resource-efficient real-time inference, multi-resolution spatiotemporal learning, and emerging human–AI collaborative paradigms. By consolidating these methodological developments, this survey aims to provide a structured reference point for ongoing research on Transformer models for time-series machine learning.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Time Series Analysis and Forecasting"],"details":null,"provenance":"openalex","attribution":"Abstract of \"A survey of transformer networks for time series forecasting\" by Jingyuan Zhao, Fulin Chu, Lili Xie et al., https://doi.org/10.1016/j.cosrev.2025.100883, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.1002/wcms.70049","kind":"article","title":"Explainable Artificial Intelligence in Drug Discovery: Bridging Predictive Power and Mechanistic Insight","authors":["Antonio Lavecchia"],"author_count":1,"year":2025,"published":"2025-09-01","venue":"Wiley Interdisciplinary Reviews Computational Molecular Science","cited_by":53,"doi":"10.1002/wcms.70049","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1002/wcms.70049","archive_url":"https://space-kitty.com/papers/doi:10.1002/wcms.70049","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W4414570715","pdf_url":"https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/wcms.70049","abstract":"ABSTRACT Explainable artificial intelligence (XAI) is increasingly essential in drug discovery, where interpretability and trust must accompany predictive accuracy. As deep learning models, particularly, deep neural networks (DNNs) and graph neural networks (GNNs), enhance molecular property prediction, de novo design, and toxicity estimation, transparent, mechanistically meaningful insights become critical. This article classifies major XAI strategies in computational molecular science, including gradient‐based attribution, perturbation analysis, surrogate modeling, counterfactual reasoning, and self‐explaining architectures. Molecular representations, such as fingerprints, SMILES, molecular graphs, and latent embeddings, are evaluated for their impact on explanation fidelity. An evaluation framework is outlined using metrics like fidelity, stability, completeness, sparsity, and usability, with emphasis on integration into drug discovery workflows. The discussion also highlights emerging directions, including neuro‐symbolic systems and physics‐informed networks that embed mechanistic constraints into statistical models. By aligning algorithmic transparency with pharmacological reasoning, XAI not only demystifies black‐box models but also supports scientific insight, regulatory compliance, and ethical AI deployment in pharmaceutical research. This article is categorized under: Data Science > Artificial Intelligence/Machine Learning Data Science > Chemoinformatics Structure and Mechanism > Computational Biochemistry and Biophysics","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Explainable Artificial Intelligence (XAI)"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Explainable Artificial Intelligence in Drug Discovery: Bridging Predictive Power and Mechanistic Insight\" by Antonio Lavecchia, https://doi.org/10.1002/wcms.70049, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.1016/j.rser.2021.111748","kind":"article","title":"Artificial neural networks for bio-based chemical production or biorefining: A review","authors":["Brett Pomeroy","Miha Grilc","Blaž Likozar"],"author_count":3,"year":2021,"published":"2021-10-22","venue":"Renewable and Sustainable Energy Reviews","cited_by":71,"doi":"10.1016/j.rser.2021.111748","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1016/j.rser.2021.111748","archive_url":"https://space-kitty.com/papers/doi:10.1016/j.rser.2021.111748","updated":null,"pmid":null,"pmcid":null,"openalex_id":"W3208932511","pdf_url":null,"abstract":null,"abstract_license":null,"license":null,"is_open_access":false,"peer_reviewed":null,"categories":["Neural Networks and Applications"],"details":null,"provenance":"openalex","attribution":"Metadata and citation counts: OpenAlex (CC0)."},{"id":"doi:10.1371/journal.pcbi.1003947","kind":"article","title":"Dynamics on Networks: The Role of Local Dynamics and Global Networks on the Emergence of Hypersynchronous Neural Activity","authors":["HELMUT J. SCHMIDT","George H. Petkov","Mark Philip Richardson","John R. Terry"],"author_count":4,"year":2014,"published":"2014-11-13","venue":"PLoS Computational Biology","cited_by":78,"doi":"10.1371/journal.pcbi.1003947","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1371/journal.pcbi.1003947","archive_url":"https://space-kitty.com/papers/doi:10.1371/journal.pcbi.1003947","updated":null,"pmid":"25393751","pmcid":null,"openalex_id":"W2073413866","pdf_url":"https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1003947&type=printable","abstract":"Graph theory has evolved into a useful tool for studying complex brain networks inferred from a variety of measures of neural activity, including fMRI, DTI, MEG and EEG. In the study of neurological disorders, recent work has discovered differences in the structure of graphs inferred from patient and control cohorts. However, most of these studies pursue a purely observational approach; identifying correlations between properties of graphs and the cohort which they describe, without consideration of the underlying mechanisms. To move beyond this necessitates the development of computational modeling approaches to appropriately interpret network interactions and the alterations in brain dynamics they permit, which in the field of complexity sciences is known as dynamics on networks. In this study we describe the development and application of this framework using modular networks of Kuramoto oscillators. We use this framework to understand functional networks inferred from resting state EEG recordings of a cohort of 35 adults with heterogeneous idiopathic generalized epilepsies and 40 healthy adult controls. Taking emergent synchrony across the global network as a proxy for seizures, our study finds that the critical strength of coupling required to synchronize the global network is significantly decreased for the epilepsy cohort for functional networks inferred from both theta (3-6 Hz) and low-alpha (6-9 Hz) bands. We further identify left frontal regions as a potential driver of seizure activity within these networks. We also explore the ability of our method to identify individuals with epilepsy, observing up to 80% predictive power through use of receiver operating characteristic analysis. Collectively these findings demonstrate that a computer model based analysis of routine clinical EEG provides significant additional information beyond standard clinical interpretation, which should ultimately enable a more appropriate mechanistic stratification of people with epilepsy leading to improved diagnostics and therapeutics.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Functional Brain Connectivity Studies"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Dynamics on Networks: The Role of Local Dynamics and Global Networks on the Emergence of Hypersynchronous Neural Activity\" by HELMUT J. SCHMIDT, George H. Petkov, Mark Philip Richardson et al., https://doi.org/10.1371/journal.pcbi.1003947, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"arxiv:2309.08600","kind":"preprint","title":"Sparse Autoencoders Find Highly Interpretable Features in Language Models","authors":["Hoagy Cunningham","Aidan Ewart","Logan Riggs","Robert Huben","Lee Sharkey"],"author_count":5,"year":2023,"published":"2023-09-15","venue":null,"cited_by":61,"doi":null,"arxiv_id":"2309.08600","nct_id":null,"source_url":"https://arxiv.org/abs/2309.08600","archive_url":"https://space-kitty.com/papers/arxiv:2309.08600","updated":"2023-10-04","pmid":null,"pmcid":null,"openalex_id":"W4386839891","pdf_url":"https://arxiv.org/pdf/2309.08600","abstract":"One of the roadblocks to a better understanding of neural networks' internals is \\textit{polysemanticity}, where neurons appear to activate in multiple, semantically distinct contexts. Polysemanticity prevents us from identifying concise, human-understandable explanations for what neural networks are doing internally. One hypothesised cause of polysemanticity is \\textit{superposition}, where neural networks represent more features than they have neurons by assigning features to an overcomplete set of directions in activation space, rather than to individual neurons. Here, we attempt to identify those directions, using sparse autoencoders to reconstruct the internal activations of a language model. These autoencoders learn sets of sparsely activating features that are more interpretable and monosemantic than directions identified by alternative approaches, where interpretability is measured by automated methods. Moreover, we show that with our learned set of features, we can pinpoint the features that are causally responsible for counterfactual behaviour on the indirect object identification task \\citep{wang2022interpretability} to a finer degree than previous decompositions. This work indicates that it is possible to resolve superposition in language models using a scalable, unsupervised method. Our method may serve as a foundation for future mechanistic interpretability work, which we hope will enable greater model transparency and steerability.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":"cc-by","is_open_access":true,"peer_reviewed":false,"categories":["cs.LG","cs.CL"],"details":null,"provenance":"arxiv+openalex","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv."},{"id":"doi:10.3389/fimmu.2022.1025330","kind":"article","title":"Ensemble deep learning enhanced with self-attention for predicting immunotherapeutic responses to cancers","authors":["Wenyi Jin","Qian Yang","Hao Chi","Kongyuan Wei","Pengpeng Zhang","Guodong Zhao","Shi Chen","Zhijia Xia","Xiaosong Li"],"author_count":9,"year":2022,"published":"2022-12-01","venue":"Frontiers in Immunology","cited_by":64,"doi":"10.3389/fimmu.2022.1025330","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.3389/fimmu.2022.1025330","archive_url":"https://space-kitty.com/papers/doi:10.3389/fimmu.2022.1025330","updated":null,"pmid":"36532083","pmcid":null,"openalex_id":"W4311069821","pdf_url":"https://www.frontiersin.org/articles/10.3389/fimmu.2022.1025330/pdf","abstract":"Introduction: Despite the many benefits immunotherapy has brought to patients with different cancers, its clinical applications and improvements are still hindered by drug resistance. Fostering a reliable approach to identifying sufferers who are sensitive to certain immunotherapeutic agents is of great clinical relevance. Methods: We propose an ELISE (Ensemble Learning for Immunotherapeutic Response Evaluation) pipeline to generate a robust and highly accurate approach to predicting individual responses to immunotherapies. ELISE employed iterative univariable logistic regression to select genetic features of patients, using Monte Carlo Tree Search (MCTS) to tune hyperparameters. In each trial, ELISE selected multiple models for integration based on add or concatenate stacking strategies, including deep neural network, automatic feature interaction learning via self-attentive neural networks, deep factorization machine, compressed interaction network, and linear neural network, then adopted the best trial to generate a final approach. SHapley Additive exPlanations (SHAP) algorithm was applied to interpret ELISE, which was then validated in an independent test set. Result: Regarding prediction of responses to atezolizumab within esophageal adenocarcinoma (EAC) patients, ELISE demonstrated a superior accuracy (Area Under Curve [AUC] = 100.00%). AC005786.3 (Mean [|SHAP value|] = 0.0097) was distinguished as the most valuable contributor to ELISE output, followed by SNORD3D (0.0092), RN7SKP72 (0.0081), EREG (0.0069), IGHV4-80 (0.0063), and MIR4526 (0.0063). Mechanistically, immunoglobulin complex, immunoglobulin production, adaptive immune response, antigen binding and others, were downregulated in ELISE-neg EAC subtypes and resulted in unfavorable responses. More encouragingly, ELISE could be extended to accurately estimate the responsiveness of various immunotherapeutic agents against other cancers, including PD1/PD-L1 suppressor against metastatic urothelial cancer (AUC = 88.86%), and MAGE-A3 immunotherapy against metastatic melanoma (AUC = 100.00%). Discussion: This study presented deep insights into integrating ensemble deep learning with self-attention as a mechanism for predicting immunotherapy responses to human cancers, highlighting ELISE as a potential tool to generate reliable approaches to individualized treatment.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Cancer Immunotherapy and Biomarkers"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Ensemble deep learning enhanced with self-attention for predicting immunotherapeutic responses to cancers\" by Wenyi Jin, Qian Yang, Hao Chi et al., https://doi.org/10.3389/fimmu.2022.1025330, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"doi:10.1007/s10822-020-00310-4","kind":"article","title":"Revealing cytotoxic substructures in molecules using deep learning","authors":["Henry Emanuel Webel","Talia B. Kimber","Silke Radetzki","Martin Neuenschwander","Marc Nazaré","Andrea Volkamer"],"author_count":6,"year":2020,"published":"2020-04-15","venue":"Journal of Computer-Aided Molecular Design","cited_by":66,"doi":"10.1007/s10822-020-00310-4","arxiv_id":null,"nct_id":null,"source_url":"https://doi.org/10.1007/s10822-020-00310-4","archive_url":"https://space-kitty.com/papers/doi:10.1007/s10822-020-00310-4","updated":null,"pmid":"32297073","pmcid":null,"openalex_id":"W3016297257","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10822-020-00310-4.pdf","abstract":"In drug development, late stage toxicity issues of a compound are the main cause of failure in clinical trials. In silico methods are therefore of high importance to guide the early design process to reduce time, costs and animal testing. Technical advances and the ever growing amount of available toxicity data enabled machine learning, especially neural networks, to impact the field of predictive toxicology. In this study, cytotoxicity prediction, one of the earliest handles in drug discovery, is investigated using a deep learning approach trained on a highly consistent in-house data set of over 34,000 compounds with a share of less than 5% of cytotoxic molecules. The model reached a balanced accuracy of over 70%, similar to previously reported studies using Random Forest. Albeit yielding good results, neural networks are often described as a black box lacking deeper mechanistic understanding of the underlying model. To overcome this absence of interpretability, a Deep Taylor Decomposition method is investigated to identify substructures that may be responsible for the cytotoxic effects, the so-called toxicophores. Furthermore, this study introduces cytotoxicity maps which provide a visual structural interpretation of the relevance of these substructures. Using this approach could be helpful in drug development to predict the potential toxicity of a compound as well as to generate new insights into the toxic mechanism. Moreover, it could also help to de-risk and optimize compounds.","abstract_license":"CC-BY (publisher license, via OpenAlex)","license":"cc-by","is_open_access":true,"peer_reviewed":null,"categories":["Computational Drug Discovery Methods"],"details":null,"provenance":"openalex","attribution":"Abstract of \"Revealing cytotoxic substructures in molecules using deep learning\" by Henry Emanuel Webel, Talia B. Kimber, Silke Radetzki et al., https://doi.org/10.1007/s10822-020-00310-4, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified."},{"id":"arxiv:2310.03652","kind":"article","title":"Extreme sparsification of physics-augmented neural networks for interpretable model discovery in mechanics","authors":["Jan N. Fuhg","Reese E. Jones","Nikolaos Bouklas"],"author_count":3,"year":2024,"published":"2024-04-16","venue":"Computer Methods in Applied Mechanics and Engineering","cited_by":53,"doi":"10.1016/j.cma.2024.116973","arxiv_id":"2310.03652","nct_id":null,"source_url":"https://doi.org/10.1016/j.cma.2024.116973","archive_url":"https://space-kitty.com/papers/arxiv:2310.03652","updated":"2023-10-05","pmid":null,"pmcid":null,"openalex_id":"W4394866287","pdf_url":"https://arxiv.org/pdf/2310.03652","abstract":"Data-driven constitutive modeling with neural networks has received increased interest in recent years due to its ability to easily incorporate physical and mechanistic constraints and to overcome the challenging and time-consuming task of formulating phenomenological constitutive laws that can accurately capture the observed material response. However, even though neural network-based constitutive laws have been shown to generalize proficiently, the generated representations are not easily interpretable due to their high number of trainable parameters. Sparse regression approaches exist that allow to obtaining interpretable expressions, but the user is tasked with creating a library of model forms which by construction limits their expressiveness to the functional forms provided in the libraries. In this work, we propose to train regularized physics-augmented neural network-based constitutive models utilizing a smoothed version of $L^{0}$-regularization. This aims to maintain the trustworthiness inherited by the physical constraints, but also enables interpretability which has not been possible thus far on any type of machine learning-based constitutive model where model forms were not assumed a-priory but were actually discovered. During the training process, the network simultaneously fits the training data and penalizes the number of active parameters, while also ensuring constitutive constraints such as thermodynamic consistency. We show that the method can reliably obtain interpretable and trustworthy constitutive models for compressible and incompressible hyperelasticity, yield functions, and hardening models for elastoplasticity, for synthetic and experimental data.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":null,"is_open_access":true,"peer_reviewed":null,"categories":["Model Reduction and Neural Networks"],"details":null,"provenance":"openalex+arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv."}],"next":"https://space-kitty.com/api/v1/topics/interpretability/papers?sort=top&limit=25&offset=25"}