Oxaliplatin(OXA),a critical third‐generation platinum chemotherapeutic,is significantly limited by suboptimal loading capacity and encapsulation efficiency in nanoparticle‐based delivery systems.To address this,we d...Oxaliplatin(OXA),a critical third‐generation platinum chemotherapeutic,is significantly limited by suboptimal loading capacity and encapsulation efficiency in nanoparticle‐based delivery systems.To address this,we developed an integrated machine learning(ML)and multi‐objective optimization(MOO)framework for the simultaneous prediction and exploration of loading efficiency(LE)and encapsulation efficiency(EE).Ensemble learning models,trained on a curated dataset of 70 experimentally characterized nanocarrier formulations,demonstrated robust predictive performance under stringent leave‐one‐paper‐out(LOPO)cross‐validation(R2=0.87 for LE,R2=0.84 for EE).The multi‐objective exploration identified a Pareto‐optimal design space,with predicted performance reaching up to 45.3%LE and 87.2%EE,and pinpointed a balanced knee‐point formulation at 40.2%LE and 83.7%EE.Interpretable ML analysis revealed surface area‐to‐volume ratio,coordination site availability,and zeta potential as the primary physicochemical drivers of OXA loading and retention.Consequently,an optimized nanocarrier profile,characterized by a particle size of 90-110 nm,a negative surface charge,and a carboxylate‐rich composition,was derived.This study establishes a predictive,data‐driven computational framework that bridges the gap between single‐objective prediction and the holistic design of high‐performance nanocarriers,providing a rational blueprint for accelerating the development of more effective OXA‐based nanotherapies for colorectal cancer.展开更多
The high-temperature performance of Co-based superalloys is primarily dictated by the coarsening kinetics and volume fraction of theγ′phase.To simultaneously optimize these two interrelated microstructural parameter...The high-temperature performance of Co-based superalloys is primarily dictated by the coarsening kinetics and volume fraction of theγ′phase.To simultaneously optimize these two interrelated microstructural parameters,we propose a dual-objective design framework that integrates explainable machine learning(XML),multi-fidelity data augmentation,and SHapley Additive exPlanations(SHAP)-based interpretability.Forγ′phase coarsening rate constant(Kr),a small experimental dataset was expanded using medium-fidelity simulations and further balanced with low-fidelity synthetic samples.Forγ′volume fraction(Vγ′),synthetic oversampling was applied to a larger dataset to mitigate distribution imbalance.ML models trained on these augmented datasets achieved high predictive accuracy,with SHAP analysis providing interpretable insights.Guided by these insights,several new compositions were proposed and validated.The optimal composition,Co-30Ni-10Al-3Ti-4Ta-5Cr-2Mo-1V(at.%),achieves a low Kr of 0.756±0.06 nm2·s-1and a high Vγ′of exceeding 70%at 1000°C,while also fulfills multiple other critical design criteria,offering a promising route for next-generation Co-based superalloys.展开更多
We present a multi-objective Bayesian active learning strategy,which greatly accelerates the discovery of super high-strength and high-ductility lead-free solder alloys.The active learning strategy demonstrates that a...We present a multi-objective Bayesian active learning strategy,which greatly accelerates the discovery of super high-strength and high-ductility lead-free solder alloys.The active learning strategy demonstrates that a machine learning model will have high generalizability if experimental data uncertainty is included,which greatly improves the model prediction or the material design accuracy.The feature-point-start forward method in multi-objective optimization adopts two Gaussian process regression(GPR)models,one for strength and one for elongation,and their outputs build up the acquisition-function-modified objective space of strength and elongation.Then,Bayesian sampling is applied to design the next experiments by balancing exploitation and exploration.Seven multi-objective active learning iterations discovered two novel super high-strength and high-ductility lead-free solder alloys.After that,various material characterizations were conducted on the two novel solder alloys,and the results exhibited their high performances in melting properties,wettability,electrical conductivity,and shear strength of the solder joint and explored the mechanism of high strength and high ductility of the alloys.The present work systematically analyzes the important role of experimental uncertainty in machine learning,especially in the global optimization for material design,which demands high generalizability of predictions.展开更多
Human skin sensory system,featuring a sophisticated threedimensional(3D)distribution of mechanoreceptors within the skin,possesses an exceptional ability to perceive a diverse range of external mechanical stimuli and ...Human skin sensory system,featuring a sophisticated threedimensional(3D)distribution of mechanoreceptors within the skin,possesses an exceptional ability to perceive a diverse range of external mechanical stimuli and accurately recognize object attributes[1].展开更多
High-entropy alloys(HEAs)have emerged as promising candidates for catalyst applications due to their inherent compositional,structural,and site-level diversities,which enable highly tunable catalytic properties.Howeve...High-entropy alloys(HEAs)have emerged as promising candidates for catalyst applications due to their inherent compositional,structural,and site-level diversities,which enable highly tunable catalytic properties.However,these complexities pose grand challenges for traditional“trial-and-error”experimentation or computationally expensive“brute-force”ab initio calculations.Machine learning(ML)demonstrates great potential to address these challenges by establishing efficient,scalable mappings from composition,structure or site environment to HEA properties.Among these properties,adsorption energy,which quantifies the binding strength between catalytic intermediates and surface sites,is a crucial indicator of catalytic activity.This review provides a comprehensive overview ofML-driven strategies for adsorption energy prediction in the context of HEAs.Two primary strategies are introduced:“direct”prediction from unrelaxed structure and“iterative”prediction viaML potential-guided relaxation modeling.Both strategies can leverage handcrafted features or end-toend frameworks such as graph neural networks.We also discuss how pretrained models on largescale databases can extend to out-of-domain HEA systems.Beyond methodology,we address key challenges and future directions,including benchmarking ML strategies,developing HEA-specific datasets,pretraining and fine-tuning,integrating chained ML models,advancing multi-objective optimization,and bridgingMLpredictions with experimental validation.By critically evaluating existing strategies and highlighting emerging trends,this review underscores the critical role of ML in advancing adsorption energy predictions,offering a foundation for accelerating the discovery and optimization of HEA catalysts.展开更多
This paper focuses on multi-modal Information Perception(IP)for Soft Robotic Hands(SRHs)using Machine Learning(ML)algorithms.A flexible Optical Fiber-based Curvature Sensor(OFCS)is fabricated,consisting of a Light-Emi...This paper focuses on multi-modal Information Perception(IP)for Soft Robotic Hands(SRHs)using Machine Learning(ML)algorithms.A flexible Optical Fiber-based Curvature Sensor(OFCS)is fabricated,consisting of a Light-Emitting Diode(LED),photosensitive detector,and optical fiber.Bending the roughened optical fiber generates lower light intensity,which reflecting the curvature of the soft finger.Together with the curvature and pressure information,multi-modal IP is performed to improve the recognition accuracy.Recognitions of gesture,object shape,size,and weight are implemented with multiple ML approaches,including the Supervised Learning Algorithms(SLAs)of K-Nearest Neighbor(KNN),Support Vector Machine(SVM),Logistic Regression(LR),and the unSupervised Learning Algorithm(un-SLA)of K-Means Clustering(KMC).Moreover,Optical Sensor Information(OSI),Pressure Sensor Information(PSI),and Double-Sensor Information(DSI)are adopted to compare the recognition accuracies.The experiment results demonstrate that the proposed sensors and recognition approaches are feasible and effective.The recognition accuracies obtained using the above ML algorithms and three modes of sensor information are higer than 85 percent for almost all combinations.Moreover,DSI is more accurate when compared to single modal sensor information and the KNN algorithm with a DSI outperforms the other combinations in recognition accuracy.展开更多
Machine learning potentials(MLPs)have become an indispensable tool in large-scale atomistic simulations.However,mostMLPs today are trained on data computed using relatively cheap density functional theory(DFT)methods ...Machine learning potentials(MLPs)have become an indispensable tool in large-scale atomistic simulations.However,mostMLPs today are trained on data computed using relatively cheap density functional theory(DFT)methods such as the Perdew-Burke-Ernzerhof(PBE)generalized gradient approximation(GGA)functional.While meta-GGAs such as the strongly constrained and appropriately normed(SCAN)functional have been shown to yield significantly improved descriptions of atomic interactions for diversely bonded systems,their higher computational cost remains an impediment to their use in MLP development.In this work,we outline a data-efficient multi-fidelity approach to constructing Materials 3-body Graph Network(M3GNet)interatomic potentials that integrate different levels of theory within a singlemodel.Using silicon and water as examples,we show that a multi-fidelity M3GNet model trained on a combined dataset of low-fidelityGGAcalculations with 10%of high-fidelity SCAN calculations can achieve accuracies comparable to a single-fidelity M3GNet model trained on a dataset comprising 8×the number of SCAN calculations.This work provides a pathway to the development of high-fidelity MLPs in a cost-effective manner by leveraging existing low-fidelity datasets.展开更多
基金financial support from the Spanish Ministry of Science and Innovation through the Ramón y Cajal Fellowship(Ayuda RYC2023‐042668‐I financiada por MICIU/AEI/10.13039/501100011033 y por el FSE+)Declaration of generative AI use:Generative AI,specifically ChatGPT(GPT‐5,OpenAI),was used to assist in editing this manuscript to improve clarity and grammar.
摘要Oxaliplatin(OXA),a critical third‐generation platinum chemotherapeutic,is significantly limited by suboptimal loading capacity and encapsulation efficiency in nanoparticle‐based delivery systems.To address this,we developed an integrated machine learning(ML)and multi‐objective optimization(MOO)framework for the simultaneous prediction and exploration of loading efficiency(LE)and encapsulation efficiency(EE).Ensemble learning models,trained on a curated dataset of 70 experimentally characterized nanocarrier formulations,demonstrated robust predictive performance under stringent leave‐one‐paper‐out(LOPO)cross‐validation(R2=0.87 for LE,R2=0.84 for EE).The multi‐objective exploration identified a Pareto‐optimal design space,with predicted performance reaching up to 45.3%LE and 87.2%EE,and pinpointed a balanced knee‐point formulation at 40.2%LE and 83.7%EE.Interpretable ML analysis revealed surface area‐to‐volume ratio,coordination site availability,and zeta potential as the primary physicochemical drivers of OXA loading and retention.Consequently,an optimized nanocarrier profile,characterized by a particle size of 90-110 nm,a negative surface charge,and a carboxylate‐rich composition,was derived.This study establishes a predictive,data‐driven computational framework that bridges the gap between single‐objective prediction and the holistic design of high‐performance nanocarriers,providing a rational blueprint for accelerating the development of more effective OXA‐based nanotherapies for colorectal cancer.
基金supports from the National Natural Science Foundation of China (Grant Nos. 52471004, 52171107, 52201203, 52401015)the Industry-University-Research Cooperation Project of Hebei Based Universities and Shijiazhuang City (Grant No. 241791237A) are gratefully acknowledged. We also greatly appreciate Dr. Bing Zhang from Yanshan University and Dr. Chun-He Chu from Henan University of Science and Technology for his insightful proposition and valuable guidance.
摘要The high-temperature performance of Co-based superalloys is primarily dictated by the coarsening kinetics and volume fraction of theγ′phase.To simultaneously optimize these two interrelated microstructural parameters,we propose a dual-objective design framework that integrates explainable machine learning(XML),multi-fidelity data augmentation,and SHapley Additive exPlanations(SHAP)-based interpretability.Forγ′phase coarsening rate constant(Kr),a small experimental dataset was expanded using medium-fidelity simulations and further balanced with low-fidelity synthetic samples.Forγ′volume fraction(Vγ′),synthetic oversampling was applied to a larger dataset to mitigate distribution imbalance.ML models trained on these augmented datasets achieved high predictive accuracy,with SHAP analysis providing interpretable insights.Guided by these insights,several new compositions were proposed and validated.The optimal composition,Co-30Ni-10Al-3Ti-4Ta-5Cr-2Mo-1V(at.%),achieves a low Kr of 0.756±0.06 nm2·s-1and a high Vγ′of exceeding 70%at 1000°C,while also fulfills multiple other critical design criteria,offering a promising route for next-generation Co-based superalloys.
基金sponsored by the Shanghai Pujiang Program(Grant no.20PJ1403700)the Guangzhou-HKUST(GZ)Joint Funding Program(nos.2023A03J0003 and 2023A03J0103)the Opening Project Fund of Materials Service Safety Assessment Facilities(MSAF-2024-107).
摘要We present a multi-objective Bayesian active learning strategy,which greatly accelerates the discovery of super high-strength and high-ductility lead-free solder alloys.The active learning strategy demonstrates that a machine learning model will have high generalizability if experimental data uncertainty is included,which greatly improves the model prediction or the material design accuracy.The feature-point-start forward method in multi-objective optimization adopts two Gaussian process regression(GPR)models,one for strength and one for elongation,and their outputs build up the acquisition-function-modified objective space of strength and elongation.Then,Bayesian sampling is applied to design the next experiments by balancing exploitation and exploration.Seven multi-objective active learning iterations discovered two novel super high-strength and high-ductility lead-free solder alloys.After that,various material characterizations were conducted on the two novel solder alloys,and the results exhibited their high performances in melting properties,wettability,electrical conductivity,and shear strength of the solder joint and explored the mechanism of high strength and high ductility of the alloys.The present work systematically analyzes the important role of experimental uncertainty in machine learning,especially in the global optimization for material design,which demands high generalizability of predictions.
摘要Human skin sensory system,featuring a sophisticated threedimensional(3D)distribution of mechanoreceptors within the skin,possesses an exceptional ability to perceive a diverse range of external mechanical stimuli and accurately recognize object attributes[1].
基金supported by the National Natural Science Foundation of China(12474189)National Key R&D Programof China(2021YFA1202300)+2 种基金Foundation of the President of China Academy of Engineering Physics(YZJJZQ2023016)Sichuan Provincial Distinguished Young Scholars Project(2025NSFJQ0022)National Natural Science Foundation of China(52394163,52371223,52101255,12192284).
摘要High-entropy alloys(HEAs)have emerged as promising candidates for catalyst applications due to their inherent compositional,structural,and site-level diversities,which enable highly tunable catalytic properties.However,these complexities pose grand challenges for traditional“trial-and-error”experimentation or computationally expensive“brute-force”ab initio calculations.Machine learning(ML)demonstrates great potential to address these challenges by establishing efficient,scalable mappings from composition,structure or site environment to HEA properties.Among these properties,adsorption energy,which quantifies the binding strength between catalytic intermediates and surface sites,is a crucial indicator of catalytic activity.This review provides a comprehensive overview ofML-driven strategies for adsorption energy prediction in the context of HEAs.Two primary strategies are introduced:“direct”prediction from unrelaxed structure and“iterative”prediction viaML potential-guided relaxation modeling.Both strategies can leverage handcrafted features or end-toend frameworks such as graph neural networks.We also discuss how pretrained models on largescale databases can extend to out-of-domain HEA systems.Beyond methodology,we address key challenges and future directions,including benchmarking ML strategies,developing HEA-specific datasets,pretraining and fine-tuning,integrating chained ML models,advancing multi-objective optimization,and bridgingMLpredictions with experimental validation.By critically evaluating existing strategies and highlighting emerging trends,this review underscores the critical role of ML in advancing adsorption energy predictions,offering a foundation for accelerating the discovery and optimization of HEA catalysts.
基金support provided by the National Natural Science Foundation of China (Nos. 61803267 and 61572328)the China Postdoctoral Science Foundation (No.2017M622757)+1 种基金the Beijing Science and Technology program (No.Z171100000817007)the National Science Foundation of China (NSFC) and the German Re-search Foundation (DFG) in the project Cross Modal Learning,NSFC 61621136008/DFG TRR-169
摘要This paper focuses on multi-modal Information Perception(IP)for Soft Robotic Hands(SRHs)using Machine Learning(ML)algorithms.A flexible Optical Fiber-based Curvature Sensor(OFCS)is fabricated,consisting of a Light-Emitting Diode(LED),photosensitive detector,and optical fiber.Bending the roughened optical fiber generates lower light intensity,which reflecting the curvature of the soft finger.Together with the curvature and pressure information,multi-modal IP is performed to improve the recognition accuracy.Recognitions of gesture,object shape,size,and weight are implemented with multiple ML approaches,including the Supervised Learning Algorithms(SLAs)of K-Nearest Neighbor(KNN),Support Vector Machine(SVM),Logistic Regression(LR),and the unSupervised Learning Algorithm(un-SLA)of K-Means Clustering(KMC).Moreover,Optical Sensor Information(OSI),Pressure Sensor Information(PSI),and Double-Sensor Information(DSI)are adopted to compare the recognition accuracies.The experiment results demonstrate that the proposed sensors and recognition approaches are feasible and effective.The recognition accuracies obtained using the above ML algorithms and three modes of sensor information are higer than 85 percent for almost all combinations.Moreover,DSI is more accurate when compared to single modal sensor information and the KNN algorithm with a DSI outperforms the other combinations in recognition accuracy.
基金ntellectually led by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division under contract No. DE-AC02-05-CH11231 (Materials Project program KC23MP)This research used resources of the National Energy Research Scientific Computing Center (NERSC), a Department of Energy Office of Science User Facility using NERSC award DOE-ERCAP0026371the support of the Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship, a Schmidt Futures program.
摘要Machine learning potentials(MLPs)have become an indispensable tool in large-scale atomistic simulations.However,mostMLPs today are trained on data computed using relatively cheap density functional theory(DFT)methods such as the Perdew-Burke-Ernzerhof(PBE)generalized gradient approximation(GGA)functional.While meta-GGAs such as the strongly constrained and appropriately normed(SCAN)functional have been shown to yield significantly improved descriptions of atomic interactions for diversely bonded systems,their higher computational cost remains an impediment to their use in MLP development.In this work,we outline a data-efficient multi-fidelity approach to constructing Materials 3-body Graph Network(M3GNet)interatomic potentials that integrate different levels of theory within a singlemodel.Using silicon and water as examples,we show that a multi-fidelity M3GNet model trained on a combined dataset of low-fidelityGGAcalculations with 10%of high-fidelity SCAN calculations can achieve accuracies comparable to a single-fidelity M3GNet model trained on a dataset comprising 8×the number of SCAN calculations.This work provides a pathway to the development of high-fidelity MLPs in a cost-effective manner by leveraging existing low-fidelity datasets.