Developing machine learning frameworks with predictive power,interpretability,and transferability is crucial,yet it faces challenges in the field of electrocatalysis.To achieve this,we employed rigorous feature engine...Developing machine learning frameworks with predictive power,interpretability,and transferability is crucial,yet it faces challenges in the field of electrocatalysis.To achieve this,we employed rigorous feature engineering to establish a finely tuned gradient boosting regressor(GBR)model,which adeptly captures the physical complexity from feature space to target variables.We demonstrated that environmental electron effects and atomic number significantly govern the success of the mapping process via global and local explanations.The finely tuned GBR model exhibits exceptional robustness in predicting CO adsorption energies(R_(ave)^(2)=0.937,RMSE=0.153 eV).Moreover,the model demonstrated remarkable transfer learning ability,showing excellent predictive power for OH,NO,and N_(2) adsorption.Importantly,the GBR model exhibits exceptional predictive capability across an extensive search space,thereby demonstrating profound adaptability and versatility.Our research framework significantly enhances the interpretability and transferability of machine learning in electrocatalysis,offering vital insights for further advancements.展开更多
Transfer-based Adversarial Attacks(TAAs)can deceive a victim model even without prior knowledge.This is achieved by leveraging the property of adversarial examples.That is,when generated from a surrogate model,they re...Transfer-based Adversarial Attacks(TAAs)can deceive a victim model even without prior knowledge.This is achieved by leveraging the property of adversarial examples.That is,when generated from a surrogate model,they retain their features if applied to other models due to their good transferability.However,adversarial examples often exhibit overfitting,as they are tailored to exploit the particular architecture and feature representation of source models.Consequently,when attempting black-box transfer attacks on different target models,their effectiveness is decreased.To solve this problem,this study proposes an approach based on a Regularized Constrained Feature Layer(RCFL).The proposed method first uses regularization constraints to attenuate the initial examples of low-frequency components.Perturbations are then added to a pre-specified layer of the source model using the back-propagation technique,in order to modify the original adversarial examples.Afterward,a regularized loss function is used to enhance the black-box transferability between different target models.The proposed method is finally tested on the ImageNet,CIFAR-100,and Stanford Car datasets with various target models,The obtained results demonstrate that it achieves a significantly higher transfer-based adversarial attack success rate compared with baseline techniques.展开更多
Artificial intelligence has potential for forecasting reactor conditions in the nuclear industry.Owing to economic and security concerns,a common method is to train data generated by simulators.However,achieving a sat...Artificial intelligence has potential for forecasting reactor conditions in the nuclear industry.Owing to economic and security concerns,a common method is to train data generated by simulators.However,achieving a satisfactory performance in practical applications is difficult because simulators imperfectly emulate reality.To bridge this gap,we propose a novel framework called simulation-to-reality domain adaptation(SRDA)for forecasting the operating parameters of nuclear reactors.The SRDA model employs a transformer-based feature extractor to capture dynamic characteristics and temporal dependencies.A parameter predictor with an improved logarithmic loss function is specifically designed to adapt to varying reactor powers.To fuse prior reactor knowledge from simulations with reality,the domain discriminator utilizes an adversarial strategy to ensure the learning of deep domain-invariant features,and the multiple kernel maximum mean discrepancy minimizes their discrepancies.Experiments on neutron fluxes and temperatures from a pressurized water reactor illustrate that the SRDA model surpasses various advanced methods in terms of predictive performance.This study is the first to use domain adaptation for real-world reactor prediction and presents a feasible solution for enhancing the transferability and generalizability of simulated data.展开更多
[Objective] The research aimed to study the transferability of Camellia sinensis EST-SSRs in Theaceae plant.[Method] Seven pairs of EST-SSRs primers which derived from Camellia sinensis EST sequence were used to ampli...[Objective] The research aimed to study the transferability of Camellia sinensis EST-SSRs in Theaceae plant.[Method] Seven pairs of EST-SSRs primers which derived from Camellia sinensis EST sequence were used to amplify the nineteen materials of Theaceae plant.[Result] Five pairs in the seven pairs of EST-SSRs primers could effectively amplify the nineteen tested varieties,and the transferability rate was 71.43%.The amplification rate of Camellia retiacalate FengShanCha,Camellia japouica CaiXia and Camellia retiacalate JuBan was the highest.The amplification rate of Camellia synaptica Sealy and Adinandra sagonica var.wallichiana（oc）Ming was the lowest.Moreover,four pairs in the five pairs of primers which could effectively amplify showed the rich polymorphism whose difference was obvious in the tested materials.[Conclusion] SSR primers which were developed from Camellia sinensis genome had the higher transferability in the different genus and species of Theaceae plant,could be used in the comparative genome research and analysis mark research of Theaceae plant.展开更多
Simple sequence repeat （SSR） or microsatellite marker is a valuable tool for several purposes, such as mapping, fingerprinting, and breeding. In the present study, an intersimple sequence repeat （ISSR）-PCR techniq...Simple sequence repeat （SSR） or microsatellite marker is a valuable tool for several purposes, such as mapping, fingerprinting, and breeding. In the present study, an intersimple sequence repeat （ISSR）-PCR technique was applied for developing SSR markers in non-heading Chinese cabbage （Brassica rapa）. A total of 190 SSRs were obtained. Among these, AG or CT （54.7%） was the most frequent repeat, followed by AC or GT （31.6%） of the microsatellites. The average number of the SSRs length array was 16 and 10 times, respectively. Based on the determined SSR sequences, 143 SSR primer pairs were designed to evaluate their transferabilities among the related species of Brassica. The number of alleles produced per marker averaged 2.91, and the polymorphism information content （PIC） value ranged from 0 to 0.863 with an average of 0.540. Monomorphism was observed in 16 primer pairs. The transferability percentage in CC genome was higher than in BB genome. More loci occurred in the BBCC genome. This result supported the hypothesis that BB genome was divergent from A and C genomes, and AA and CC genomes were relatively close. The polymorphic primers can be exploited for further evolution, fingerprinting, and variety identification.展开更多
Currently development of new marker types has shifted from anonymous DNA fragments to gene-based markers. Simple Sequence Repeats (SSRs) are useful DNA markers in plant genetic research including in peanut. However, d...Currently development of new marker types has shifted from anonymous DNA fragments to gene-based markers. Simple Sequence Repeats (SSRs) are useful DNA markers in plant genetic research including in peanut. However, de novo development of SSRs is expensive and time consuming. Gene-based DNA markers are transferable among related species owing to the conserved nature of genes. In this study transferability of sorghum EST-SSR (SbEST-SSR) markers to peanut was prospected. A set of 411 SbEST-SSR primer pairs were used to amplify peanut genomic DNA extracted from cultivated peanut where 39% of them successfully amplified. A comparison of amplification patterns between sorghum and peanut showed similar banding pattern with majority of transferable SbEST-SSRs. Among these transferable SSR markers, 14% have detected polymorphism among 4 resistant and 4 susceptible peanut lines for rust and late leaf spot diseases. These transferable markers will benefit peanut genome research by not only providing additional DNA markers for population genetic analyses, but also allowing comparative mapping to be possible between peanut and sorghum—a possible monocot-dicot comparison.展开更多
We characterized 14 anonymous nuclear loci from Pinus thunbergii Parl., an important pine species native to Japan. One hundred and twenty-six single nucleotide polymorphisms （SNPs） were identified from these loci, g...We characterized 14 anonymous nuclear loci from Pinus thunbergii Parl., an important pine species native to Japan. One hundred and twenty-six single nucleotide polymorphisms （SNPs） were identified from these loci, giving a frequency of 1 SNP per 51 bp. Nucleotide di- versity （0） ranged from 1.06 × 10^-3 to 11.87 × 10^-3, with all average of 4.99 × 10^-3. Only one locus （mK45） deviated significantly from the Hardy-Weinberg equilibrium. Thirteen of 14 loci were applicable in other pine species. These loci will be useful for nucleotide variation studies and will provide material for SNP-based marker development in P. thun- bergii and related species.展开更多
Glacier mass balance, the difference between accumulation and ablation at the glacier surface, is the direct reflection of the local climate regime. Under global warming, the simulation of glacier mass balance at the ...Glacier mass balance, the difference between accumulation and ablation at the glacier surface, is the direct reflection of the local climate regime. Under global warming, the simulation of glacier mass balance at the regional scale has attracted increasing interests. This study selects Urumqi Glacier No. 1 as the testbed for examining the transferability in space and time of two commonly used glacier mass balance simulation models: i.e., the Degree-Day Model(DDM) and the simplified Energy Balance Model(s EBM). Four experiments were carried out for assessing both models’ temporal and spatial transferability. The results show that the spatial transferability of both the DDM and s EBM is strong, whereas the temporal transferability of the DDM is relatively weak. For all four experiments, the overall simulation effect of the s EBM is better than that of the DDM. At the zone around Equilibrium Line Altitude(ELA), the DDM performed better than the s EBM.Also, the accuracy of parameters, including the lapse rate of air temperature and vertical gradient of precipitation at the glacier surface, is of great significance for improving the spatial transferability of both models.展开更多
The current adversarial attacks against deep learning models have achieved incredible success in the white-box scenario.However,they often exhibit weak transferability in the black-box scenario,especially when attacki...The current adversarial attacks against deep learning models have achieved incredible success in the white-box scenario.However,they often exhibit weak transferability in the black-box scenario,especially when attacking those with defense mechanisms.In this work,we propose a new transfer-based blackbox attack called the channel decomposition attack method(CDAM).It can attack multiple black-box models by enhancing the transferability of the adversarial examples.On the one hand,it tunes the gradient and stabilizes the update direction by decomposing the channels of the input example and calculating the aggregate gradient.On the other hand,it helps to escape from local optima by initializing the data point with random noise.Besides,it could combine with other transfer-based attacks flexibly.Extensive experiments on the standard ImageNet dataset show that our method could significantly improve the transferability of adversarial attacks.Compared with the state-of-the-art method,our approach improves the average success rate from 88.2%to 96.6%when attacking three adversarially trained black-box models,demonstrating the remaining shortcomings of existing deep learning models.展开更多
Transferability of five nuclear microsatellite markers (Jc-16, Jc-31, Jc-32, Jc-35 and Jc-37) that were originally developed for J. communis was tested to J. procera. Jc-31 & Jc-37 showed successful amplifications...Transferability of five nuclear microsatellite markers (Jc-16, Jc-31, Jc-32, Jc-35 and Jc-37) that were originally developed for J. communis was tested to J. procera. Jc-31 & Jc-37 showed successful amplifications and polymorphism in J. procera. Jc-35 which had been reported as polymorphic in J. communis was monomorphic in J. procera while the primer pair for Jc-32 failed to record any amplification. The remaining one primer pair (Jc-16) showed double loci ampli-fication in both J. procera and the control J. communis suggesting further examination of the primer pair and its binding sites. Genetic variation of six Ethiopian J. procera populations: Chilimo, Goba, Menagesha-Suba, Wef-Washa, Yabelo and Ziquala was assessed based on the two polymorphic loci (Jc-31 & Jc-37) in 20 - 24 individuals of each population. From these two loci, a total of 41 alleles could be retrieved. Two populations that are located south east of the Great Rift Valley together harboured 75% of private alleles signifying their deviant geo-ecological zones and suggesting special consideration for conservation. Chilimo, which is at the western margin of Juniper habitat in Ethiopian central highlands scored the highest fixation (FIS = 0.584) entailing lower immigrant genes and hence higher inbreeding. The AMOVA revealed that 97% of the variation resided within the?population while still among population variation was significant展开更多
As households are equipped with smart meters,supervised Machine Learning(ML)models and especially Non-Intrusive Load Monitoring(NILM)disaggregation algorithms are becoming increasingly important.To be robust,these mod...As households are equipped with smart meters,supervised Machine Learning(ML)models and especially Non-Intrusive Load Monitoring(NILM)disaggregation algorithms are becoming increasingly important.To be robust,these models require a large amount of data,which is difficult to collect.Consequently,the generation of meaningful synthetic data is becoming more relevant.We use a simulation framework to generate multiple datasets using different techniques and evaluate their quality statistically by measuring the performance of NILM models for transferability.We demonstrate that the method of data generation is crucial to train ML models in a meaningful way.The experiments conducted reveal that adding noise to the synthetic smart meter data is essential to train robust NILM models for transferability.The best results are obtained when this noise is derived from unknown appliances for which no ground truth data is available.Since we observed that NILM models can provide unstable results,we develop a reliable evaluation methodology,based on Cochran’s sample size.Finally,we compare the quality of the generated synthetic data with real data and observe that multiple NILM models trained on synthetic data perform significantly better than those trained on real data.展开更多
Safety performance functions(SPFs),or crash-prediction models,have played an important role in identifying the factors contributing to crashes,predicting crash counts and identifying hotspots.Since a great deal of tim...Safety performance functions(SPFs),or crash-prediction models,have played an important role in identifying the factors contributing to crashes,predicting crash counts and identifying hotspots.Since a great deal of time and effort is needed to estimate an SPF,previous studies have sought to determine the transferability of particular SPFs;that is,the extent to which they can be applied to data from other regions.Although many efforts have been made to examine micro-level SPF transferability,few studies have focused on macro-level SPF transferability.There has been little transferability analysis of macro-level SPFs in the international context,especially between western countries.This study therefore evaluates the transferability of SPFs for several states in the USA(Illinois,Florida and Colorado)and for Italy.The SPFs were developed using data from counties in the United States and provincias in Italy,and the results revealed multiple common significant variables between the two countries.Transferability indexes were then calculated between the SPFs.These showed that the Italy SPFs for total crashes and bicycle crashes were transferable to US data after calibration factors were applied,whereas the US SPFs for total and bicycle crashes,with the exception of the Colorado SPF,could not be transferred to the Italian data.On the other hand,none of the pedestrian SPFs developed was transferable to other countries.This paper provides insights into the applicability of macro-level SPFs between the USA and Italy,and shows a good potential for international SPF transferability.Nevertheless,further investigation is needed of SPF transferability between a wider range of countries.展开更多
In recent years,universal adversarial per-turbation(UAP)has attracted the attention of many re-searchers due to its good generalization.However,in order to generate an appropriate UAP,current methods usually require e...In recent years,universal adversarial per-turbation(UAP)has attracted the attention of many re-searchers due to its good generalization.However,in order to generate an appropriate UAP,current methods usually require either accessing the original dataset or meticulously constructing optimization functions and proxy datasets.In this paper,we aim to elimi-nate any dependency on proxy datasets and explore a method for generating Universal Adversarial Pertur-bations(UAP)on a single image.After revisiting re-search on UAP,we discovered that the key to gener-ating UAP lies in the accumulation of Individual Ad-versarial Perturbation(IAP)gradient,which prompted us to study the method of accumulating gradients from an IAP.We designed a simple and effective process to generate UAP,which only includes three steps:pre-cessing,generating an IAP and scaling the perturba-tions.Through our proposed process,any IAP gener-ated on an image can be constructed into a UAP with comparable performance,indicating that UAP can be generated free of data.Extensive experiments on var-ious classifiers and attack approaches demonstrate the superiority of our method on efficiency and aggressiveness.展开更多
Deep neural networks remain susceptible to adversarial examples,where the goal of an adversarial attack is to introduce small perturbations to the original examples in order to confuse the model without being easily d...Deep neural networks remain susceptible to adversarial examples,where the goal of an adversarial attack is to introduce small perturbations to the original examples in order to confuse the model without being easily detected.Although many adversarial attack methods produce adversarial examples that have achieved great results in the whitebox setting,they exhibit low transferability in the black-box setting.In order to improve the transferability along the baseline of the gradient-based attack technique,we present a novel Stochastic Gradient Accumulation Momentum Iterative Attack(SAMI-FGSM)in this study.In particular,during each iteration,the gradient information is calculated using a normal sampling approach that randomly samples around the sample points,with the highest probability of capturing adversarial features.Meanwhile,the accumulated information of the sampled gradient from the previous iteration is further considered to modify the current updated gradient,and the original gradient attack direction is changed to ensure that the updated gradient direction is more stable.Comprehensive experiments conducted on the ImageNet dataset show that our method outperforms existing state-of-the-art gradient-based attack techniques,achieving an average improvement of 10.2%in transferability.展开更多
Bay-site carboxyl functionalized perylene diimide derivative 1,7-COOH-PDI-C_(12)(PDI-COOH)was synthesized and distinct enhanced fluorescence was observed through combining with calcium ion(Ca^(2+))in THF/H_(2)O soluti...Bay-site carboxyl functionalized perylene diimide derivative 1,7-COOH-PDI-C_(12)(PDI-COOH)was synthesized and distinct enhanced fluorescence was observed through combining with calcium ion(Ca^(2+))in THF/H_(2)O solution.The assembly and fluorescence behavior of PDI-COOH/Ca^(2+)were studied in detail by changing hydration state with different concentrations.Based on the differences in assembly morphology and stoichiometric ratios of PDICOOH/Ca^(2+),we proposed the fluorescence emission mechanism of PDI-COOH/Ca^(2+)in THF/H_(2)O and THF,respectively.This work reveals a novel strategy of aggregated state fluorescence enhancement and reminds us of the important role of water in molecular fluorescence emission and assembly.展开更多
Full-color imaging is essential in digital pathology for accurate tissue analysis.Utilizing advanced optical modulation and phase retrieval algorithms,Fourier ptychographic microscopy(FPM)offers a powerful solution fo...Full-color imaging is essential in digital pathology for accurate tissue analysis.Utilizing advanced optical modulation and phase retrieval algorithms,Fourier ptychographic microscopy(FPM)offers a powerful solution for high-throughput digital pathology,combining high resolution,large field of view,and extended depth of field(DOF).However,the full-color capabilities of FPM are hindered by coherent color artifacts and reduced computational efficiency,which significantly limits its practical applications.Color-transferbased FPM(CFPM)has emerged as a potential solution,theoretically reducing both acquisition and reconstruction threefold time.Yet,existing methods fall short of achieving the desired reconstruction speed and colorization quality.In this study,we report a generalized dual-color-space constrained model for FPM colorization.This model provides a mathematical framework for model-based FPM colorization,enabling a closed-form solution without the need for redundant iterative calculations.Our approach,termed generalized CFPM(gCFPM),achieves colorization within seconds for megapixel-scale images,delivering superior colorization quality in terms of both colorfulness and sharpness,along with an extended DOF.Both simulations and experiments demonstrate that gCFPM surpasses state-of-the-art methods across all evaluated criteria.Our work offers a robust and comprehensive workflow for high-throughput full-color pathological imaging using FPM platforms,laying a solid foundation for future advancements in methodology and engineering.展开更多
Objective Poxviruses are zoonotic pathogens that infect humans,mammals,vertebrates,and arthropods.However,the specific role of ticks in transmission and evolution of these viruses remains unclear.Methods Transcriptomi...Objective Poxviruses are zoonotic pathogens that infect humans,mammals,vertebrates,and arthropods.However,the specific role of ticks in transmission and evolution of these viruses remains unclear.Methods Transcriptomic and metatranscriptomic raw data from 329 sampling pools of seven tick species across five continents were mined to assess the diversity and abundance of poxviruses.Chordopoxviral sequences were assembled and subjected to phylogenetic analysis to trace the origins of the unblasted fragments within these sequences.Results Fifty-eight poxvirus species,representing two subfamilies and 20 genera,were identified,with 212 poxviral sequences assembled.A substantial proportion of AT-rich fragments were detected in the assembled poxviral genomes.These genomic sequences contained fragments originating from rodents,archaea,and arthropods.Conclusion Our findings indicate that ticks play a significant role in the transmission and evolution of poxviruses.These viruses demonstrate the capacity to modulate virulence and adaptability through horizontal gene transfer,gene recombination,and gene mutations,thereby promoting co-existence and co-evolution with their hosts.This study advances understanding of the ecological dynamics of poxvirus transmission and evolution and highlights the potential role of ticks as vectors and vessels in these processes.展开更多
This article presents a detailed theoretical hybrid analysis of the magnetism and the thermal radiative heat transfer in the presence of heat generation affecting the behavior of the dispersed gold nanoparticles(AuNPs...This article presents a detailed theoretical hybrid analysis of the magnetism and the thermal radiative heat transfer in the presence of heat generation affecting the behavior of the dispersed gold nanoparticles(AuNPs)through the blood vessels of the human body.The rheology of gold-blood nanofluid is treated as magnetohydrodynamic(MHD)flow with ferromagnetic properties.The AuNPs take different shapes as bricks,cylinders,and platelets which are considered in changing the nanofluid flow behavior.Physiologically,the blood is circulated under the kinetics of the peristaltic action.The mixed properties of the slip flow,the gravity,the space porosity,the transverse ferromagnetic field,the thermal radiation,the nanoparticles shape factors,the peristaltic amplitude ratio,and the concentration of the AuNPs are interacted and analyzed for the gold-blood circulation in the inclined tube.The appropriate model for the thermal conductivity of the nanofluid is chosen to be the effective Hamilton-Crosser model.The undertaken nanofluid can be treated as incompressible non-Newtonian ferromagnetic fluid.The solutions of the partial differential governing equations of the MHD nanofluid flow are executed by the strategy of perturbation approach under the assumption of long wavelength and low Reynolds number.Graphs for the streamwise velocity distributions,temperature distributions,pressure gradients,pressure drops,and streamlines are presented under the influences of the pertinent properties.The practical implementation of this research finds application in treating cancer through a technique known as photothermal therapy(PTT).The results indicate the control role of the magnetism,the heat generation,the shape factors of the AuNPs,and its concentration on the enhancement of the thermal properties and the streamwise velocity of the nanofluid.The results reveal a marked enhancement in the temperature profiles of the nanofluid,prominently influenced by both the intensified heat source and the heightened volume fractions of the nanoparticles.Furthermore,the platelet shape is regarded as most advantageous for heat conduction owing to its highest effective thermal conductivity.AuNPs proved strong efficiency in delivering and targeting the drug to reach the affected area with tumors.These results offer valuable insights into evaluating the effectiveness of PTT in addressing diverse cancer conditions and regulating their progression.展开更多
BACKGROUND Hepatocellular carcinoma(HCC)is a prevalent and life-threatening cancer with increasing incidence worldwide.High Ki-67 risk stratification is closely associated with higher recurrence rates and worse outcom...BACKGROUND Hepatocellular carcinoma(HCC)is a prevalent and life-threatening cancer with increasing incidence worldwide.High Ki-67 risk stratification is closely associated with higher recurrence rates and worse outcomes following curative therapies in patients with HCC.However,the performance of radiomic and deep transfer learning(DTL)models derived from biparametric magnetic resonance imaging(bpMRI)in predicting Ki-67 risk stratification and recurrence-free survival(RFS)in patients with HCC remains limited.AIM To develop a nomogram model integrating bpMRI-based radiomic and DTL signatures for predicting Ki-67 risk stratification and RFS in patients with HCC.METHODS This study included 198 patients with histopathologically confirmed HCC who underwent preoperative bpMRI.Ki-67 risk stratification was categorized as high(>20%)or low(≤20%)according to immunohistochemical staining.Radiomic and DTL signatures were extracted from the T2-weighted and arterial-phase images and combined through a random forest algorithm to establish radiomic and DTL models,respectively.Multivariate regression analysis identified clinical risk factors for high Ki-67 risk stratification,and a predictive nomogram model was developed.RESULTS A nonsmooth margin and the absence of an enhanced capsule were independent factors for high Ki-67 risk stratification.The area under the curve(AUC)of the clinical model was 0.77,while those of the radiomic and DTL models were 0.81 and 0.87,respectively,for the prediction of high Ki-67 risk stratification,and the nomogram model achieved a better AUC of 0.92.The median RFS times for patients with high and low Ki-67 risk stratification were 33.00 months and 66.73 months,respectively(P<0.001).Additionally,patients who were predicted to have high Ki-67 risk stratification by the nomogram model had a lower median RFS than those who were predicted to have low Ki-67 risk stratification(33.53 vs 66.74 months,P=0.007).CONCLUSION Our developed nomogram model demonstrated good performance in predicting Ki-67 risk stratification and predicting survival outcomes in patients with HCC.展开更多
In scenarios such as vehicle radiation monitoring and unmanned aerial vehicle radiation detection,rapid measurements using a NaI(Tl)detector often result in low photon counts,weak characteristic peaks,and significant ...In scenarios such as vehicle radiation monitoring and unmanned aerial vehicle radiation detection,rapid measurements using a NaI(Tl)detector often result in low photon counts,weak characteristic peaks,and significant statistical fluctuations.These issues can lead to potential failures in peak-searching-based identification methods.To address the low precision associated with short-duration measurements of radionuclides,this paper proposes an identification algorithm that leverages heterogeneous spectral transfer to develop a low-count energy spectral identification model.Comparative experiments demonstrated that transferring samples from 26 classes of simulated heterogeneous gamma spectra aids in creating a reliable model for measured gamma spectra.With only 10%of target domain samples used for training,the accuracy on real low-count spectral samples was 95.56%.This performance shows a significant improvement over widely employed full-spectrum analysis methods trained on target domain samples.The proposed method also exhibits strong generalization capabilities,effectively mitigating overfitting issues in low-count energy spectral classification under short-duration measurements.展开更多
基金supported by the Research Grants Council of Hong Kong(CityU 11305919 and 11308620)and NSFC/RGC Joint Research Scheme N_CityU104/19Hong Kong Research Grant Council Collaborative Research Fund:C1002-21G and C1017-22Gsupported by the Hong Kong Research Grant Council Collaborative Research Fund:C6021-19E.
文摘Developing machine learning frameworks with predictive power,interpretability,and transferability is crucial,yet it faces challenges in the field of electrocatalysis.To achieve this,we employed rigorous feature engineering to establish a finely tuned gradient boosting regressor(GBR)model,which adeptly captures the physical complexity from feature space to target variables.We demonstrated that environmental electron effects and atomic number significantly govern the success of the mapping process via global and local explanations.The finely tuned GBR model exhibits exceptional robustness in predicting CO adsorption energies(R_(ave)^(2)=0.937,RMSE=0.153 eV).Moreover,the model demonstrated remarkable transfer learning ability,showing excellent predictive power for OH,NO,and N_(2) adsorption.Importantly,the GBR model exhibits exceptional predictive capability across an extensive search space,thereby demonstrating profound adaptability and versatility.Our research framework significantly enhances the interpretability and transferability of machine learning in electrocatalysis,offering vital insights for further advancements.
基金supported by the Intelligent Policing Key Laboratory of Sichuan Province(No.ZNJW2022KFZD002)This work was supported by the Scientific and Technological Research Program of Chongqing Municipal Education Commission(Grant Nos.KJQN202302403,KJQN202303111).
文摘Transfer-based Adversarial Attacks(TAAs)can deceive a victim model even without prior knowledge.This is achieved by leveraging the property of adversarial examples.That is,when generated from a surrogate model,they retain their features if applied to other models due to their good transferability.However,adversarial examples often exhibit overfitting,as they are tailored to exploit the particular architecture and feature representation of source models.Consequently,when attempting black-box transfer attacks on different target models,their effectiveness is decreased.To solve this problem,this study proposes an approach based on a Regularized Constrained Feature Layer(RCFL).The proposed method first uses regularization constraints to attenuate the initial examples of low-frequency components.Perturbations are then added to a pre-specified layer of the source model using the back-propagation technique,in order to modify the original adversarial examples.Afterward,a regularized loss function is used to enhance the black-box transferability between different target models.The proposed method is finally tested on the ImageNet,CIFAR-100,and Stanford Car datasets with various target models,The obtained results demonstrate that it achieves a significantly higher transfer-based adversarial attack success rate compared with baseline techniques.
基金supported by the Industry-University Cooperation Project in Fujian Province University(No.2023H6006)the State Key Laboratory of Reliability and Intelligence of Electrical Equipment(No.EERI-KF20200005)。
文摘Artificial intelligence has potential for forecasting reactor conditions in the nuclear industry.Owing to economic and security concerns,a common method is to train data generated by simulators.However,achieving a satisfactory performance in practical applications is difficult because simulators imperfectly emulate reality.To bridge this gap,we propose a novel framework called simulation-to-reality domain adaptation(SRDA)for forecasting the operating parameters of nuclear reactors.The SRDA model employs a transformer-based feature extractor to capture dynamic characteristics and temporal dependencies.A parameter predictor with an improved logarithmic loss function is specifically designed to adapt to varying reactor powers.To fuse prior reactor knowledge from simulations with reality,the domain discriminator utilizes an adversarial strategy to ensure the learning of deep domain-invariant features,and the multiple kernel maximum mean discrepancy minimizes their discrepancies.Experiments on neutron fluxes and temperatures from a pressurized water reactor illustrate that the SRDA model surpasses various advanced methods in terms of predictive performance.This study is the first to use domain adaptation for real-world reactor prediction and presents a feasible solution for enhancing the transferability and generalizability of simulated data.
基金Supported by"Provincial and Ministerial Key Discipline,Provincial Key Laboratory of University and School Laboratory Sharing Platform" ItemSouthwest Forestry University Key Fund Item(110909)~~
文摘[Objective] The research aimed to study the transferability of Camellia sinensis EST-SSRs in Theaceae plant.[Method] Seven pairs of EST-SSRs primers which derived from Camellia sinensis EST sequence were used to amplify the nineteen materials of Theaceae plant.[Result] Five pairs in the seven pairs of EST-SSRs primers could effectively amplify the nineteen tested varieties,and the transferability rate was 71.43%.The amplification rate of Camellia retiacalate FengShanCha,Camellia japouica CaiXia and Camellia retiacalate JuBan was the highest.The amplification rate of Camellia synaptica Sealy and Adinandra sagonica var.wallichiana（oc）Ming was the lowest.Moreover,four pairs in the five pairs of primers which could effectively amplify showed the rich polymorphism whose difference was obvious in the tested materials.[Conclusion] SSR primers which were developed from Camellia sinensis genome had the higher transferability in the different genus and species of Theaceae plant,could be used in the comparative genome research and analysis mark research of Theaceae plant.
文摘Simple sequence repeat （SSR） or microsatellite marker is a valuable tool for several purposes, such as mapping, fingerprinting, and breeding. In the present study, an intersimple sequence repeat （ISSR）-PCR technique was applied for developing SSR markers in non-heading Chinese cabbage （Brassica rapa）. A total of 190 SSRs were obtained. Among these, AG or CT （54.7%） was the most frequent repeat, followed by AC or GT （31.6%） of the microsatellites. The average number of the SSRs length array was 16 and 10 times, respectively. Based on the determined SSR sequences, 143 SSR primer pairs were designed to evaluate their transferabilities among the related species of Brassica. The number of alleles produced per marker averaged 2.91, and the polymorphism information content （PIC） value ranged from 0 to 0.863 with an average of 0.540. Monomorphism was observed in 16 primer pairs. The transferability percentage in CC genome was higher than in BB genome. More loci occurred in the BBCC genome. This result supported the hypothesis that BB genome was divergent from A and C genomes, and AA and CC genomes were relatively close. The polymorphic primers can be exploited for further evolution, fingerprinting, and variety identification.
文摘Currently development of new marker types has shifted from anonymous DNA fragments to gene-based markers. Simple Sequence Repeats (SSRs) are useful DNA markers in plant genetic research including in peanut. However, de novo development of SSRs is expensive and time consuming. Gene-based DNA markers are transferable among related species owing to the conserved nature of genes. In this study transferability of sorghum EST-SSR (SbEST-SSR) markers to peanut was prospected. A set of 411 SbEST-SSR primer pairs were used to amplify peanut genomic DNA extracted from cultivated peanut where 39% of them successfully amplified. A comparison of amplification patterns between sorghum and peanut showed similar banding pattern with majority of transferable SbEST-SSRs. Among these transferable SSR markers, 14% have detected polymorphism among 4 resistant and 4 susceptible peanut lines for rust and late leaf spot diseases. These transferable markers will benefit peanut genome research by not only providing additional DNA markers for population genetic analyses, but also allowing comparative mapping to be possible between peanut and sorghum—a possible monocot-dicot comparison.
文摘We characterized 14 anonymous nuclear loci from Pinus thunbergii Parl., an important pine species native to Japan. One hundred and twenty-six single nucleotide polymorphisms （SNPs） were identified from these loci, giving a frequency of 1 SNP per 51 bp. Nucleotide di- versity （0） ranged from 1.06 × 10^-3 to 11.87 × 10^-3, with all average of 4.99 × 10^-3. Only one locus （mK45） deviated significantly from the Hardy-Weinberg equilibrium. Thirteen of 14 loci were applicable in other pine species. These loci will be useful for nucleotide variation studies and will provide material for SNP-based marker development in P. thun- bergii and related species.
基金supported by the Key Research Program of Frontier Sciences of Chinese Academy of Sciences(Grant No.QYZDB-SSW-SYS024)National Natural Science Foundation of China(Grant Nos.41771081 and 41761134093).
文摘Glacier mass balance, the difference between accumulation and ablation at the glacier surface, is the direct reflection of the local climate regime. Under global warming, the simulation of glacier mass balance at the regional scale has attracted increasing interests. This study selects Urumqi Glacier No. 1 as the testbed for examining the transferability in space and time of two commonly used glacier mass balance simulation models: i.e., the Degree-Day Model(DDM) and the simplified Energy Balance Model(s EBM). Four experiments were carried out for assessing both models’ temporal and spatial transferability. The results show that the spatial transferability of both the DDM and s EBM is strong, whereas the temporal transferability of the DDM is relatively weak. For all four experiments, the overall simulation effect of the s EBM is better than that of the DDM. At the zone around Equilibrium Line Altitude(ELA), the DDM performed better than the s EBM.Also, the accuracy of parameters, including the lapse rate of air temperature and vertical gradient of precipitation at the glacier surface, is of great significance for improving the spatial transferability of both models.
基金This work was supported by Sichuan Science and Technology Program[No.2022YFG0315,2022YFG0174]Sichuan Gas Turbine Research Institute stability support project of China Aero Engine Group Co.,Ltd.[GJCZ-2019-71]Key project of Chengdu[No.2019-YF09-00044-CG].
文摘The current adversarial attacks against deep learning models have achieved incredible success in the white-box scenario.However,they often exhibit weak transferability in the black-box scenario,especially when attacking those with defense mechanisms.In this work,we propose a new transfer-based blackbox attack called the channel decomposition attack method(CDAM).It can attack multiple black-box models by enhancing the transferability of the adversarial examples.On the one hand,it tunes the gradient and stabilizes the update direction by decomposing the channels of the input example and calculating the aggregate gradient.On the other hand,it helps to escape from local optima by initializing the data point with random noise.Besides,it could combine with other transfer-based attacks flexibly.Extensive experiments on the standard ImageNet dataset show that our method could significantly improve the transferability of adversarial attacks.Compared with the state-of-the-art method,our approach improves the average success rate from 88.2%to 96.6%when attacking three adversarially trained black-box models,demonstrating the remaining shortcomings of existing deep learning models.
基金The German Academic Exchange Service(DAAD)is appreciated for financial support to the first author
文摘Transferability of five nuclear microsatellite markers (Jc-16, Jc-31, Jc-32, Jc-35 and Jc-37) that were originally developed for J. communis was tested to J. procera. Jc-31 & Jc-37 showed successful amplifications and polymorphism in J. procera. Jc-35 which had been reported as polymorphic in J. communis was monomorphic in J. procera while the primer pair for Jc-32 failed to record any amplification. The remaining one primer pair (Jc-16) showed double loci ampli-fication in both J. procera and the control J. communis suggesting further examination of the primer pair and its binding sites. Genetic variation of six Ethiopian J. procera populations: Chilimo, Goba, Menagesha-Suba, Wef-Washa, Yabelo and Ziquala was assessed based on the two polymorphic loci (Jc-31 & Jc-37) in 20 - 24 individuals of each population. From these two loci, a total of 41 alleles could be retrieved. Two populations that are located south east of the Great Rift Valley together harboured 75% of private alleles signifying their deviant geo-ecological zones and suggesting special consideration for conservation. Chilimo, which is at the western margin of Juniper habitat in Ethiopian central highlands scored the highest fixation (FIS = 0.584) entailing lower immigrant genes and hence higher inbreeding. The AMOVA revealed that 97% of the variation resided within the?population while still among population variation was significant
基金funded by the German Ministry for Economics Affairs and Climate Action(BMWK)within the project ForeSightNEXT.
文摘As households are equipped with smart meters,supervised Machine Learning(ML)models and especially Non-Intrusive Load Monitoring(NILM)disaggregation algorithms are becoming increasingly important.To be robust,these models require a large amount of data,which is difficult to collect.Consequently,the generation of meaningful synthetic data is becoming more relevant.We use a simulation framework to generate multiple datasets using different techniques and evaluate their quality statistically by measuring the performance of NILM models for transferability.We demonstrate that the method of data generation is crucial to train ML models in a meaningful way.The experiments conducted reveal that adding noise to the synthetic smart meter data is essential to train robust NILM models for transferability.The best results are obtained when this noise is derived from unknown appliances for which no ground truth data is available.Since we observed that NILM models can provide unstable results,we develop a reliable evaluation methodology,based on Cochran’s sample size.Finally,we compare the quality of the generated synthetic data with real data and observe that multiple NILM models trained on synthetic data perform significantly better than those trained on real data.
文摘Safety performance functions(SPFs),or crash-prediction models,have played an important role in identifying the factors contributing to crashes,predicting crash counts and identifying hotspots.Since a great deal of time and effort is needed to estimate an SPF,previous studies have sought to determine the transferability of particular SPFs;that is,the extent to which they can be applied to data from other regions.Although many efforts have been made to examine micro-level SPF transferability,few studies have focused on macro-level SPF transferability.There has been little transferability analysis of macro-level SPFs in the international context,especially between western countries.This study therefore evaluates the transferability of SPFs for several states in the USA(Illinois,Florida and Colorado)and for Italy.The SPFs were developed using data from counties in the United States and provincias in Italy,and the results revealed multiple common significant variables between the two countries.Transferability indexes were then calculated between the SPFs.These showed that the Italy SPFs for total crashes and bicycle crashes were transferable to US data after calibration factors were applied,whereas the US SPFs for total and bicycle crashes,with the exception of the Colorado SPF,could not be transferred to the Italian data.On the other hand,none of the pedestrian SPFs developed was transferable to other countries.This paper provides insights into the applicability of macro-level SPFs between the USA and Italy,and shows a good potential for international SPF transferability.Nevertheless,further investigation is needed of SPF transferability between a wider range of countries.
基金supported in part by the Natural Science Foundation of China under Grant 62372395in part by the Research Foundation of Education Bureau of Hunan Province under Grant No.24A0105in part by the Postgraduate Scientific Research Innovation Project of Hunan Province(Grant No.CX20230546).
文摘In recent years,universal adversarial per-turbation(UAP)has attracted the attention of many re-searchers due to its good generalization.However,in order to generate an appropriate UAP,current methods usually require either accessing the original dataset or meticulously constructing optimization functions and proxy datasets.In this paper,we aim to elimi-nate any dependency on proxy datasets and explore a method for generating Universal Adversarial Pertur-bations(UAP)on a single image.After revisiting re-search on UAP,we discovered that the key to gener-ating UAP lies in the accumulation of Individual Ad-versarial Perturbation(IAP)gradient,which prompted us to study the method of accumulating gradients from an IAP.We designed a simple and effective process to generate UAP,which only includes three steps:pre-cessing,generating an IAP and scaling the perturba-tions.Through our proposed process,any IAP gener-ated on an image can be constructed into a UAP with comparable performance,indicating that UAP can be generated free of data.Extensive experiments on var-ious classifiers and attack approaches demonstrate the superiority of our method on efficiency and aggressiveness.
基金supported in part by the National Natural Science Foundation(62202118,U24A20241)in part by Major Scientific and Technological Special Project of Guizhou Province([2024]014,[2024]003)+1 种基金in part by Scientific and Technological Research Projects from Guizhou Education Department(Qian jiao ji[2023]003)in part by Guizhou Science and Technology Department Hundred Level Innovative Talents Project(GCC[2023]018).
文摘Deep neural networks remain susceptible to adversarial examples,where the goal of an adversarial attack is to introduce small perturbations to the original examples in order to confuse the model without being easily detected.Although many adversarial attack methods produce adversarial examples that have achieved great results in the whitebox setting,they exhibit low transferability in the black-box setting.In order to improve the transferability along the baseline of the gradient-based attack technique,we present a novel Stochastic Gradient Accumulation Momentum Iterative Attack(SAMI-FGSM)in this study.In particular,during each iteration,the gradient information is calculated using a normal sampling approach that randomly samples around the sample points,with the highest probability of capturing adversarial features.Meanwhile,the accumulated information of the sampled gradient from the previous iteration is further considered to modify the current updated gradient,and the original gradient attack direction is changed to ensure that the updated gradient direction is more stable.Comprehensive experiments conducted on the ImageNet dataset show that our method outperforms existing state-of-the-art gradient-based attack techniques,achieving an average improvement of 10.2%in transferability.
文摘Bay-site carboxyl functionalized perylene diimide derivative 1,7-COOH-PDI-C_(12)(PDI-COOH)was synthesized and distinct enhanced fluorescence was observed through combining with calcium ion(Ca^(2+))in THF/H_(2)O solution.The assembly and fluorescence behavior of PDI-COOH/Ca^(2+)were studied in detail by changing hydration state with different concentrations.Based on the differences in assembly morphology and stoichiometric ratios of PDICOOH/Ca^(2+),we proposed the fluorescence emission mechanism of PDI-COOH/Ca^(2+)in THF/H_(2)O and THF,respectively.This work reveals a novel strategy of aggregated state fluorescence enhancement and reminds us of the important role of water in molecular fluorescence emission and assembly.
基金supported by the National Natural Science Foundation of China(Grant Nos.12104500 and 82430062)the Key Research and Development Projects of Shaanxi Province(Grant No.2023-YBSF-263),the Shenzhen Engineering Research Centre(Grant No.XMHT20230115004)the Shenzhen Science and Technology Innovation Commission(Grant No.KCXFZ20201221173207022).
文摘Full-color imaging is essential in digital pathology for accurate tissue analysis.Utilizing advanced optical modulation and phase retrieval algorithms,Fourier ptychographic microscopy(FPM)offers a powerful solution for high-throughput digital pathology,combining high resolution,large field of view,and extended depth of field(DOF).However,the full-color capabilities of FPM are hindered by coherent color artifacts and reduced computational efficiency,which significantly limits its practical applications.Color-transferbased FPM(CFPM)has emerged as a potential solution,theoretically reducing both acquisition and reconstruction threefold time.Yet,existing methods fall short of achieving the desired reconstruction speed and colorization quality.In this study,we report a generalized dual-color-space constrained model for FPM colorization.This model provides a mathematical framework for model-based FPM colorization,enabling a closed-form solution without the need for redundant iterative calculations.Our approach,termed generalized CFPM(gCFPM),achieves colorization within seconds for megapixel-scale images,delivering superior colorization quality in terms of both colorfulness and sharpness,along with an extended DOF.Both simulations and experiments demonstrate that gCFPM surpasses state-of-the-art methods across all evaluated criteria.Our work offers a robust and comprehensive workflow for high-throughput full-color pathological imaging using FPM platforms,laying a solid foundation for future advancements in methodology and engineering.
基金financially supported by the Shanghai New Three-Year Action Plan for Public Health(Grant No.GWVI-11.1-03)National Natural Science Foundation of China(Grant No.81872673).
文摘Objective Poxviruses are zoonotic pathogens that infect humans,mammals,vertebrates,and arthropods.However,the specific role of ticks in transmission and evolution of these viruses remains unclear.Methods Transcriptomic and metatranscriptomic raw data from 329 sampling pools of seven tick species across five continents were mined to assess the diversity and abundance of poxviruses.Chordopoxviral sequences were assembled and subjected to phylogenetic analysis to trace the origins of the unblasted fragments within these sequences.Results Fifty-eight poxvirus species,representing two subfamilies and 20 genera,were identified,with 212 poxviral sequences assembled.A substantial proportion of AT-rich fragments were detected in the assembled poxviral genomes.These genomic sequences contained fragments originating from rodents,archaea,and arthropods.Conclusion Our findings indicate that ticks play a significant role in the transmission and evolution of poxviruses.These viruses demonstrate the capacity to modulate virulence and adaptability through horizontal gene transfer,gene recombination,and gene mutations,thereby promoting co-existence and co-evolution with their hosts.This study advances understanding of the ecological dynamics of poxvirus transmission and evolution and highlights the potential role of ticks as vectors and vessels in these processes.
文摘This article presents a detailed theoretical hybrid analysis of the magnetism and the thermal radiative heat transfer in the presence of heat generation affecting the behavior of the dispersed gold nanoparticles(AuNPs)through the blood vessels of the human body.The rheology of gold-blood nanofluid is treated as magnetohydrodynamic(MHD)flow with ferromagnetic properties.The AuNPs take different shapes as bricks,cylinders,and platelets which are considered in changing the nanofluid flow behavior.Physiologically,the blood is circulated under the kinetics of the peristaltic action.The mixed properties of the slip flow,the gravity,the space porosity,the transverse ferromagnetic field,the thermal radiation,the nanoparticles shape factors,the peristaltic amplitude ratio,and the concentration of the AuNPs are interacted and analyzed for the gold-blood circulation in the inclined tube.The appropriate model for the thermal conductivity of the nanofluid is chosen to be the effective Hamilton-Crosser model.The undertaken nanofluid can be treated as incompressible non-Newtonian ferromagnetic fluid.The solutions of the partial differential governing equations of the MHD nanofluid flow are executed by the strategy of perturbation approach under the assumption of long wavelength and low Reynolds number.Graphs for the streamwise velocity distributions,temperature distributions,pressure gradients,pressure drops,and streamlines are presented under the influences of the pertinent properties.The practical implementation of this research finds application in treating cancer through a technique known as photothermal therapy(PTT).The results indicate the control role of the magnetism,the heat generation,the shape factors of the AuNPs,and its concentration on the enhancement of the thermal properties and the streamwise velocity of the nanofluid.The results reveal a marked enhancement in the temperature profiles of the nanofluid,prominently influenced by both the intensified heat source and the heightened volume fractions of the nanoparticles.Furthermore,the platelet shape is regarded as most advantageous for heat conduction owing to its highest effective thermal conductivity.AuNPs proved strong efficiency in delivering and targeting the drug to reach the affected area with tumors.These results offer valuable insights into evaluating the effectiveness of PTT in addressing diverse cancer conditions and regulating their progression.
基金Supported by Clinical Trials from the Third Affiliated Hospital of Soochow University,No.2024-156Changzhou Science and Technology Program,No.CJ20244017。
文摘BACKGROUND Hepatocellular carcinoma(HCC)is a prevalent and life-threatening cancer with increasing incidence worldwide.High Ki-67 risk stratification is closely associated with higher recurrence rates and worse outcomes following curative therapies in patients with HCC.However,the performance of radiomic and deep transfer learning(DTL)models derived from biparametric magnetic resonance imaging(bpMRI)in predicting Ki-67 risk stratification and recurrence-free survival(RFS)in patients with HCC remains limited.AIM To develop a nomogram model integrating bpMRI-based radiomic and DTL signatures for predicting Ki-67 risk stratification and RFS in patients with HCC.METHODS This study included 198 patients with histopathologically confirmed HCC who underwent preoperative bpMRI.Ki-67 risk stratification was categorized as high(>20%)or low(≤20%)according to immunohistochemical staining.Radiomic and DTL signatures were extracted from the T2-weighted and arterial-phase images and combined through a random forest algorithm to establish radiomic and DTL models,respectively.Multivariate regression analysis identified clinical risk factors for high Ki-67 risk stratification,and a predictive nomogram model was developed.RESULTS A nonsmooth margin and the absence of an enhanced capsule were independent factors for high Ki-67 risk stratification.The area under the curve(AUC)of the clinical model was 0.77,while those of the radiomic and DTL models were 0.81 and 0.87,respectively,for the prediction of high Ki-67 risk stratification,and the nomogram model achieved a better AUC of 0.92.The median RFS times for patients with high and low Ki-67 risk stratification were 33.00 months and 66.73 months,respectively(P<0.001).Additionally,patients who were predicted to have high Ki-67 risk stratification by the nomogram model had a lower median RFS than those who were predicted to have low Ki-67 risk stratification(33.53 vs 66.74 months,P=0.007).CONCLUSION Our developed nomogram model demonstrated good performance in predicting Ki-67 risk stratification and predicting survival outcomes in patients with HCC.
基金supported by the National Defense Fundamental Research Project(No.JCKY2022404C005)the Nuclear Energy Development Project(No.23ZG6106)+1 种基金the Sichuan Scientific and Technological Achievements Transfer and Transformation Demonstration Project(No.2023ZHCG0026)the Mianyang Applied Technology Research and Development Project(No.2021ZYZF1005)。
文摘In scenarios such as vehicle radiation monitoring and unmanned aerial vehicle radiation detection,rapid measurements using a NaI(Tl)detector often result in low photon counts,weak characteristic peaks,and significant statistical fluctuations.These issues can lead to potential failures in peak-searching-based identification methods.To address the low precision associated with short-duration measurements of radionuclides,this paper proposes an identification algorithm that leverages heterogeneous spectral transfer to develop a low-count energy spectral identification model.Comparative experiments demonstrated that transferring samples from 26 classes of simulated heterogeneous gamma spectra aids in creating a reliable model for measured gamma spectra.With only 10%of target domain samples used for training,the accuracy on real low-count spectral samples was 95.56%.This performance shows a significant improvement over widely employed full-spectrum analysis methods trained on target domain samples.The proposed method also exhibits strong generalization capabilities,effectively mitigating overfitting issues in low-count energy spectral classification under short-duration measurements.
