Artificial intelligence(AI)is profoundly reshaping the discovery and design of organic light-emitting diode(OLED)materials,shifting conventional intuition-driven development into an integrated,datadriven paradigm.The ...Artificial intelligence(AI)is profoundly reshaping the discovery and design of organic light-emitting diode(OLED)materials,shifting conventional intuition-driven development into an integrated,datadriven paradigm.The increasing demand for high-performance OLED emitters with ultra-narrow emission spectrum and enhanced operational stability has highlighted the urgent need for a dedicated,multi-scale computational framework tailored to OLED-specific challenges.This review proposes a systematic AI-driven framework that combines quantum chemistry calculations,property prediction models,and generative algorithms to enable high-throughput screening and inverse design workflows for organic luminescent materials.Each component is critically analyzed in terms of theoretical underpinnings,practical benefits,inherent limitations,and avenues for further optimization.By presenting detailed case studies,we elucidate how AI approaches can tackle key bottlenecks in OLED material discovery and development.Moreover,we highlight essential future directions,including the integration of domain-specific expertise,the establishment of high-quality experimentally validated datasets,and the creation of molecular generation models specifically adapted for luminescent materials.Overall,this review aims to provide a comprehensive roadmap for advancing AI-guided materials research,offering transferable insights that extend beyond OLEDs to a broad range of organic optoelectronic materials.展开更多
Supervised learning often requires a large number of labeled examples,which has become a critical bottleneck in the case that manual annotating the class labels is costly.To mitigate this issue,a new framework called ...Supervised learning often requires a large number of labeled examples,which has become a critical bottleneck in the case that manual annotating the class labels is costly.To mitigate this issue,a new framework called pairwise comparison(Pcomp)classification is proposed to allow training examples only weakly annotated with pairwise comparison,i.e.,which one of two examples is more likely to be positive.The previous study solves Pcomp problems by minimizing the classification error,which may lead to less robust model due to its sensitivity to class distribution.In this paper,we propose a robust learning framework for Pcomp data along with a pairwise surrogate loss called Pcomp-AUC.It provides an unbiased estimator to equivalently maximize AUC without accessing the precise class labels.Theoretically,we prove the consistency with respect to AUC and further provide the estimation error bound for the proposed method.Empirical studies on multiple datasets validate the effectiveness of the proposed method.展开更多
1 Introduction.Recently,some research efforts[1]have tried to combine selfsupervised learning and active learning to reduce the cost of labeling samples.However,this method is difficult to effectively improve the mode...1 Introduction.Recently,some research efforts[1]have tried to combine selfsupervised learning and active learning to reduce the cost of labeling samples.However,this method is difficult to effectively improve the model performance because it does not consider the feature representation performance of the examples on the pretext task.In order to overcome this shortcoming,we propose a deep active sampling framework with self-supervised representation learning.展开更多
基金supported by the Beijing Natural Science Foundation(2242054)the National Natural Science Foundation of China(62075006 and 62475177)。
文摘Artificial intelligence(AI)is profoundly reshaping the discovery and design of organic light-emitting diode(OLED)materials,shifting conventional intuition-driven development into an integrated,datadriven paradigm.The increasing demand for high-performance OLED emitters with ultra-narrow emission spectrum and enhanced operational stability has highlighted the urgent need for a dedicated,multi-scale computational framework tailored to OLED-specific challenges.This review proposes a systematic AI-driven framework that combines quantum chemistry calculations,property prediction models,and generative algorithms to enable high-throughput screening and inverse design workflows for organic luminescent materials.Each component is critically analyzed in terms of theoretical underpinnings,practical benefits,inherent limitations,and avenues for further optimization.By presenting detailed case studies,we elucidate how AI approaches can tackle key bottlenecks in OLED material discovery and development.Moreover,we highlight essential future directions,including the integration of domain-specific expertise,the establishment of high-quality experimentally validated datasets,and the creation of molecular generation models specifically adapted for luminescent materials.Overall,this review aims to provide a comprehensive roadmap for advancing AI-guided materials research,offering transferable insights that extend beyond OLEDs to a broad range of organic optoelectronic materials.
基金Natural Science Foundation of Jiangsu Province,China(BK20222012,BK20211517)National Key R&D Program of China(2020AAA0107000)National Natural Science Foundation of China(Grant No.62222605)。
文摘Supervised learning often requires a large number of labeled examples,which has become a critical bottleneck in the case that manual annotating the class labels is costly.To mitigate this issue,a new framework called pairwise comparison(Pcomp)classification is proposed to allow training examples only weakly annotated with pairwise comparison,i.e.,which one of two examples is more likely to be positive.The previous study solves Pcomp problems by minimizing the classification error,which may lead to less robust model due to its sensitivity to class distribution.In this paper,we propose a robust learning framework for Pcomp data along with a pairwise surrogate loss called Pcomp-AUC.It provides an unbiased estimator to equivalently maximize AUC without accessing the precise class labels.Theoretically,we prove the consistency with respect to AUC and further provide the estimation error bound for the proposed method.Empirical studies on multiple datasets validate the effectiveness of the proposed method.
文摘1 Introduction.Recently,some research efforts[1]have tried to combine selfsupervised learning and active learning to reduce the cost of labeling samples.However,this method is difficult to effectively improve the model performance because it does not consider the feature representation performance of the examples on the pretext task.In order to overcome this shortcoming,we propose a deep active sampling framework with self-supervised representation learning.
