Search

The contrasting differences that exist in the English and Cantonese phonological systems result in divergent articulatory production between them. This article aims to investigate the negative transfer of Cantonese EFL learners in acquiring English consonant clusters and the constraint rankings of Cantonese and English syllable systems by comparing syllable structures and analyzing experimental statistics. The objective is to elucidate the reasons behind their negative transfer in English syllable acquisition. The research results demonstrated that Cantonese EFL learners found it easier to acquire consonant clusters in the onset position compared to the coda position. Additionally, an increase in consonants in the coda position posed greater difficulties for Cantonese EFL learners. The results also indicated that participants encountered the most problems with deletion, followed by substitution and epenthesis. The OT analysis reveals that the different rankings of faithfulness and markedness constraints in Cantonese and English led to transfer errors. [ABSTRACT FROM AUTHOR]

The aim of this paper is to improve the parameter estimation accuracy of the system to be identified by using measurements from a known system. By introducing the transfer gain matrix and setting the effective identification criterion, a novel transfer sparse identification method is raised, which deals with the sparse issues more precise. Besides, the unbiased form is given in the parameter analysis and the recursion form can prevent the dimension catastrophe related problems. Moreover, in order to test the effects of the transfer and avoid bad performance, a negative transfer analysis condition is carried out. Finally, the simulation verifies the enhancements and benefits of the proposed transfer sparse identification method, confirming that the transfer performance outperforms better than that of no transfer, especially on the zero parameters identification. [ABSTRACT FROM AUTHOR]

Building on the cross-linguistic variability in the meaning of vague quantifiers, this study explores the potential for negative transfer in Italian-Slovenian bilinguals concerning the use of quantificational determiners, specifically the translational equivalents of the English “many”, that is the Slovenian "precej" and "veliko". The aim is to identify relevant aspects of pragmatic knowledge for cross-linguistic influence. The study presents the results of a sentence-picture verification task in which Slovenian native speakers and Italian-Slovenian bilinguals evaluated sentences of the form "Quantifier X are Y" in relation to visual contexts. The results suggest that Italian learners of Slovenian, unlike Slovenian native speakers, fail to distinguish between "precej" and "veliko". This finding aligns with the negative transfer hypothesis. The study highlights the potential role of pragmatic knowledge in cross-linguistic transfer, particularly in the context of vague quantifiers. [ABSTRACT FROM AUTHOR]

Multi-task learning (MTL) is widely used in the online recommendation and financial services for multi-step conversion estimation, but current works often overlook the sequential dependence among tasks. In particular, sequential dependence multi-task learning (SDMTL) faces challenges in dealing with complex task correlations and extracting valuable information in real-world scenarios, leading to negative transfer and a deterioration in the performance. Herein, a systematic learning paradigm of the SDMTL problem is established for the first time, which applies to more general multi-step conversion scenarios with longer conversion paths or various task dependence relationships. Meanwhile, an SDMTL architecture, named Task-Aware Feature Extraction (TAFE), is designed to enable the dynamic task representation learning from a sample-wise view. TAFE selectively reconstructs the implicit shared information corresponding to each sample case and performs the explicit task-specific extraction under dependence constraints, which can avoid the negative transfer, resulting in more effective information sharing and joint representation learning. Extensive experiment results demonstrate the effectiveness and applicability of the proposed theoretical and implementation frameworks. Furthermore, the online evaluations at MYbank showed that TAFE had an average increase of 9.22% and 3.76% in various scenarios on the post-view click-through & conversion rate (CTCVR) estimation task. Currently, TAFE is deployed in an online platform to provide various traffic services. [ABSTRACT FROM AUTHOR]

Multitask optimization algorithms can simultaneously derive the best solution for different tasks; however, the convergence speed of such algorithms is slow when frequent negative transfers occur. This is primarily because the similarity function of different tasks is designed only in the decision or target space. Moreover, an algorithm is prone to fall into local optima when population diversity is lost. To reduce negative migration and balance diversity and enhance convergence of multitask optimization algorithms, a multitask particle swarm optimization algorithm based on dual spatial similarity (MTPSO-DSS) is developed in this study. A new similarity function is built into the algorithm for the different tasks based on both decision and target spaces, whereby the transfer probability is adaptively adjusted. The new similarity function, which is more rigorous and accurate, can reduce the probability of negative migration and maintain the convergence speed. Furthermore, a new updating method is designed to handle negative migration and increase diversity of search directions. Adaptive mutation and non-allelic gene crossover strategies are designed to increase the diversity of the algorithm and help it escape from local optima. To verify the performance of the proposed algorithm, nine general multitasking optimization test functions are tested via the proposed algorithm, and the results are compared with other eight multitasking algorithms. The proposed algorithm outperformed the other algorithms for most functions in terms of convergence accuracy and speed, and the average improvement in the convergence accuracy compared with the other eight algorithms is between 23.35 and 99.99%. [ABSTRACT FROM AUTHOR]

Recently, transfer learning has gained popularity in the machine learning community. Transfer Learning (TL) has emerged as a promising paradigm that leverages knowledge learned from one or more related domains to improve prediction accuracy in a target domain with limited data. However, for time series forecasting (TSF) applications, transfer learning is relatively new. This paper addresses the need for empirical studies as identified in recent reviews advocating the need for practical guidelines for Transfer Learning approaches and method designs for time series forecasting. The main contribution of this paper is the suggestion of a comprehensive framework for Transfer Learning Sensitivity Analysis (SA) for time series forecasting. We achieve this by identifying various parameters seen from various angles of transfer learning applied to time series, aiming to uncover factors and insights that influence the performance of transfer learning in time series forecasting. Undoubtedly, symmetry appears to be a core aspect in the consideration of these factors and insights. A further contribution is the introduction of four TL performance metrics encompassed in our framework. These TL performance metrics provide insight into the extent of the transferability between the source and the target domains. Analyzing whether the benefits of transferred knowledge are equally or unequally accessible and applicable across different domains or tasks speaks to the requirement of symmetry or asymmetry in transfer learning. Moreover, these TL performance metrics inform on the possibility of the occurrence of negative transfers and also provide insight into the possible vulnerability of the network to catastrophic forgetting. Finally, we discuss a sensitivity analysis of an Ensemble TL technique use case (with Multilayer Perceptron models) as a proof of concept to validate the suggested framework. While the results from the experiments offer empirical insights into various parameters that impact the transfer learning gain, they also raise the question of network dimensioning requirements when designing, specifically, a neural network for transfer learning. [ABSTRACT FROM AUTHOR]

Parallel corpora for low-resource Arabic dialects and English are limited and small-scale, and most neural machine translation models, including Google Translate, rely mainly on parallel corpora of standard Arabic and English to train for dialectal Arabic translation. A model well trained to translate to and from standard Arabic is believed to efficiently translate dialectal Arabic, given their similarities. This study demonstrates the impact of not using large-scale, dialect-specific parallel corpora by quantitatively and qualitatively analyzing the performance of Google Translate in translating Egyptian Arabic adjuncts. Compared to human reference translation, Google Translate achieved a low BLEU score of 14.69. Qualitative analysis showed that reliance on standard Arabic parallel corpora caused a negative transfer problem manifested in the literal translation of idiomatic adjuncts, the misinterpretation of dialectal adjuncts as main clause constituents, the translation of dialectal adjuncts after orthographically similar standard Arabic words, and the use of standard Arabic common lexical meanings to translate dialect-specific adjuncts. This study’s findings will be relevant for researchers interested in dialectal Arabic neural machine translation and has implications for investment in the development of large-scale, dialect-specific corpora to better process the peculiarities of Arabic dialects and reduce the effect of negative transfer from standard Arabic. [ABSTRACT FROM AUTHOR]

The visions of Industry 4.0 and 5.0 have reinforced the industrial environment. They have also made artificial intelligence incorporated as a major facilitator. Diagnosing machine faults has become a solid foundation for automatically recognizing machine failure, and thus timely maintenance can ensure safe operations. Transfer learning is a promising solution that can enhance the machine fault diagnosis model by borrowing pre-trained knowledge from the source model and applying it to the target model, which typically involves two datasets. In response to the availability of multiple datasets, this paper proposes using selective and adaptive incremental transfer learning (SA-ITL), which fuses three algorithms, namely, the hybrid selective algorithm, the transferability enhancement algorithm, and the incremental transfer learning algorithm. It is a selective algorithm that enables selecting and ordering appropriate datasets for transfer learning and selecting useful knowledge to avoid negative transfer. The algorithmalso adaptively adjusts the portion of training data to balance the learning rate and training time. The proposed algorithm is evaluated and analyzed using ten benchmark datasets. Compared with other algorithms fromexisting works, SA-ITL improves the accuracy of all datasets. Ablation studies present the accuracy enhancements of the SA-ITL, including the hybrid selective algorithm (1.22%--3.82%), transferability enhancement algorithm (1.91%--4.15%), and incremental transfer learning algorithm (0.605%--2.68%). These also show the benefits of enhancing the target model with heterogeneous image datasets that widen the range of domain selection between source and target domains. [ABSTRACT FROM AUTHOR]

Transfer learning is an essential tool for improving the performance of primary tasks by leveraging information from auxiliary data resources. In this work, we propose Adaptive Robust Transfer Learning (ART), a flexible pipeline of performing transfer learning with generic machine learning algorithms. We establish the nonasymptotic learning theory of ART, providing a provable theoretical guarantee for achieving adaptive transfer while preventing negative transfer. Additionally, we introduce an ART‐integrated‐aggregating machine that produces a single final model when multiple candidate algorithms are considered. We demonstrate the promising performance of ART through extensive empirical studies on regression, classification, and sparse learning. We further present a real‐data analysis for a mortality study. [ABSTRACT FROM AUTHOR]

In this work, we study the transfer learning problem under high-dimensional generalized linear models (GLMs), which aim to improve the fit on target data by borrowing information from useful source data. Given which sources to transfer, we propose a transfer learning algorithm on GLM, and derive its l 1 / l 2 -estimation error bounds as well as a bound for a prediction error measure. The theoretical analysis shows that when the target and sources are sufficiently close to each other, these bounds could be improved over those of the classical penalized estimator using only target data under mild conditions. When we don't know which sources to transfer, an algorithm-free transferable source detection approach is introduced to detect informative sources. The detection consistency is proved under the high-dimensional GLM transfer learning setting. We also propose an algorithm to construct confidence intervals of each coefficient component, and the corresponding theories are provided. Extensive simulations and a real-data experiment verify the effectiveness of our algorithms. We implement the proposed GLM transfer learning algorithms in a new R package glmtrans, which is available on CRAN. for this article are available online. [ABSTRACT FROM AUTHOR]

Copyright of Arts for Linguistic & Literary Studies is the property of Thamar University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Currently, most models rarely consider the negative transfer problem in the research field of cross-subject EEG emotion recognition. To solve this problem, this paper proposes a semi-supervised domain adaptive algorithm based on few labeled samples of target subject, which called multi-domain geodesic flow kernel dynamic distribution alignment (MGFKD). It consists of three modules: 1) GFK common feature extractor: projects the feature distribution of source and target subjects to the Grassmann manifold space, and obtains the latent common features of the two feature distributions through GFK method. 2) Source domain selector: obtains pseudo-labels of the target subject through weak classifier, finds ''golden source subjects'' by using few known labels of target subjects. 3) Label corrector: uses a dynamic distribution balance strategy to correct the pseudo-labels of the target subject. We conducted comparison experiments on the SEED and SEED-IV datasets, and the results show that MGFKD outperforms unsupervised and semi-supervised domain adaptation algorithms, achieving an average accuracy of 87.51±7.68% and 68.79±8.25% on the SEED and SEED-IV datasets with only one labeled sample per video for target subject. Especially when the number of source domains is set as 6 and the number of known labels is set as 5, the accuracy increase to 90.20±7.57% and 69.99±7.38%, respectively. The above results prove that our proposed algorithm can efficiently and quickly improve the cross-subject EEG emotion classification performance. Since it only need a small number of labeled samples of new subjects, making it has strong application value in future EEG-based emotion recognition applications.

The present study aimed to explore learners’ perceptions while using Google Translate (GT) during online English-Writing lessons and to evaluate its effectiveness on learners’ English writing skills. Online questionnaires and individual interviews were used to collect self-reported opinions from 24 Vietnamese students. Learner writing samples from the 12-week online session were also collected in order to identify possible improvements in writing. The findings revealed that the learners generally reported a positive effect on their writing from using Google Translate. In contrast, textual analysis of learners’ writing samples indicated that although there were slight improvements in their writing skills, problems still exist. According to an error analysis that was conducted, negative transfer of structures from the students’ L1 (Vietnamese) accounted for most of the lexical and syntactic errors identified. The findings demonstrate that GT is a useful support tool for teaching English writing. The contrastive analysis in the present study contributes to language interference studies and discussions on bilingual and multilingual education in the Vietnamese context.

The present study aimed at investigating the effect of the L1 negative transfer on the writing of EFL Saudi students. It examined the common grammatical errors that Saudi students made while writing in English and analyzed the sources of their errors. The samples consisted of English essays written by 74 freshmen female students enrolled at Majmaah University in Saudi Arabia. The samples were analyzed and the errors classified according to grammatical sub-categories including tenses, singular/plural markers, prepositions, articles and pronouns. This study found that 81% of the students' errors can be accounted for in terms of L1 transfer. Grammatical errors were the most frequent ones, recording 67% of the total rate of errors. More specifically, preposition (40%) and tense (32%) errors constituted the most frequent subcategories of the grammatical errors. It was concluded that learning basic linguistic differences between Saudi students' L1 and English language is a necessary condition for helping the learners to overcome the problem of first language interference in their writing.

This study explores how gender and first language (L1) conceptual transfer influence the use of English prepositions among 50 Vietnamese university students aged 18 to 25, enrolled in an English language program. The participants' comprehension of English prepositions in terms of usage, meaning, and context was assessed using pre- and post-tests consisting of 20 multiple-choice questions. The data analysis employed descriptive and inferential statistics, including chi-square tests and independent samples t-tests. Both the preand post-test results revealed a moderate level of English prepositional usage, with all participants displaying improvement. Female participants performed slightly better than males on the post-test, while there were no notable gender differences in pre-test scores. In both tests, participants with high levels of L1 conceptual transfer performed significantly poorer compared to those with low levels. These findings highlight the significant role of L1 conceptual transfer in the English prepositional usage of Vietnamese language learners, while suggesting a minor impact of gender. Further research is needed to delve deeper into this relationship. The results emphasize the importance of addressing L1 conceptual transfer in English language instruction, specifically regarding prepositions, within the context of Vietnam. [ABSTRACT FROM AUTHOR]

This study aimed to investigate the acquisition of liaison in English by Chinese-speaking learners. Ten second-year postgraduate students of non-English majors in Tongji University, China, were invited to take part in an experiment. They were asked to prepare recordings of a set of English materials, including phrases, dialogues, and a talking topic, before and after a self-study training on liaison. To examine every type of liaison in their speech, the study analysed the recordings using the speech analysis software Praat. The results showed that before the training, the students negatively transferred the native language (L1) pattern to the target second language (L2). This kind of negative transfer of L1 Chinese to the acquisition of liaison in L2 English could be explained by the differences between English and Chinese syllables. After the training, the students showed substantial improvement in phrase and dialogue reading. The findings are expected to help both teachers and students gain a better understanding of liaison and the differences between English and Chinese syllables, thus contributing to English teaching and learning.

A vast body of available literature presents controversial perspectives on the role of L1 in learning and communicating an L2. The current exploratory qualitative study attempted to investigate the reported experiences of ten Afghan EFL learners regarding their L1 (Farsi Dari) role in communicating an L2 (English). The data collected through Skype interviews with ten participants were analyzed thematically. The findings indicated that both positive and negative transfers of L1 occur at different levels in L2. The emerging themes revealed that L2 is scaffolded with L1 proficiency; further, it helps generate ideas, improve self-esteem, and reduce anxiety. The findings also indicated some L1 interferences in L2 communication particularly pronunciation, grammar, and vocabulary. The current study's findings suggest that the language instructors should be aware of the language transfers, both positive and negative, to provide quality teaching to EFL learners. Further research studies can be conducted through a different research design such as quantitative or mixed-method exploring the EFL perspectives.

Copyright of Journal of Guangdong University of Technology is the property of Journal of Guangdong University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The article explores a particular case of negative transfer of the Surrealist artistic project during the early phase of the French-centred movement within the Romanian avant-garde magazines network. Dismissed as “inauthentic” or mediated for its scandalous allure, Surrealism is unsuccessful at making inroads into the Romanian avant-garde press, closely connected to International Constructivism. This process triggers a delayed, albeit impressive, response to the Surrealist aesthetic and political thought, toward the end of the 1920s, and is, initially, largely regarded as a fraught association. Arguing that artistic circulation and transmission is not always a synchronous or symmetrical process in a centre-periphery direction, the article opens an inquiry into the motivations and implications of the Romanian negative response to early Surrealism. Additionally, the article provides insights into the complex interplay between Constructivism and Surrealism in the space of the Romanian avant-garde periodicals, essential for the understanding of Central and Eastern European avant-gardes, and a new perspective on how the periphery moderates the Surrealist label and its artistic project. [ABSTRACT FROM AUTHOR]

The typical way to conduct data-driven prognosis is to train a degradation model with historical data, then apply the model to predict failure for in-service units. Most existing works assume the historical data and in-service data are from the same process. In practice, however, different but related processes can share similar degradation patterns. Thus, the historical data from these processes are expected to provide useful prognosis information for each other. In this article, we propose a data-level transfer learning framework to extract useful and shared information from different processes to benefit the prognosis of in-service units. In this framework, the degradation data in each process is modeled by a mixed effects model. To facilitate the information sharing among different mixed effects models, a hierarchical Bayesian structure is proposed to model and connect the distributions of mixed effects in different mixed models. Because the degradation paths in different processes are rarely the same, the dimension of the mixed effects/regressor in each process can be different. To handle this issue, we propose a tailored linear transformation to marginalize or expand the distributions of mixed effects in different degradation processes to achieve consistent dimensions. The transferred information is finally incorporated with the degradation data from in-service units to conduct prognosis. The proposed method is validated and compared with various benchmarks in extensive numerical studies and two case studies. The results show the proposed method can successfully transfer useful information in different processes to benefit the prognosis. [ABSTRACT FROM AUTHOR]

Introduction: Boxing as a sport is growing on Chinese campuses, resulting in a coaching shortage. The human pose estimation technology can be employed to estimate boxing poses and teach interns to relieve the shortage. Currently, 3D cameras can provide more depth information than 2D cameras. It can potentially improve the estimation. However, the input channels are inconsistent between 2D and 3D images, and there is a lack of detailed analysis about the key point location, which indicates the network design for improving the human pose estimation technology. Method: Therefore, a model transfer with channel patching was implemented to solve the problems of channel inconsistency. The differences between the key points were analyzed. Three popular and highly structured 2D models of OpenPose (OP), stacked Hourglass (HG), and High Resolution (HR) networks were employed. Ways of reusing RGB channels were investigated to fill up the depth channel. Then, their performances were investigated to find out the limitations of each network structure. Results and discussion: The results show that model transfer learning by the mean way of RGB channels patching the lacking channel can improve the average accuracies of pose key points from 1 to 20% than without transfer. 3D accuracies are 0.3 to 0.5% higher than 2D baselines. The stacked structure of the network shows better on hip and knee points than the parallel structure, although the parallel design shows much better on the residue points. As a result, the model transfer can practically fulfill boxing pose estimation from 2D to 3D. [ABSTRACT FROM AUTHOR]

Recently, transfer learning has gained popularity in the machine learning community. Transfer Learning (TL) has emerged as a promising paradigm that leverages knowledge learned from one or more related domains to improve prediction accuracy in a target domain with limited data. However, for time series forecasting (TSF) applications, transfer learning is relatively new. This paper addresses the need for empirical studies as identified in recent reviews advocating the need for practical guidelines for Transfer Learning approaches and method designs for time series forecasting. The main contribution of this paper is the suggestion of a comprehensive framework for Transfer Learning Sensitivity Analysis (SA) for time series forecasting. We achieve this by identifying various parameters seen from various angles of transfer learning applied to time series, aiming to uncover factors and insights that influence the performance of transfer learning in time series forecasting. Undoubtedly, symmetry appears to be a core aspect in the consideration of these factors and insights. A further contribution is the introduction of four TL performance metrics encompassed in our framework. These TL performance metrics provide insight into the extent of the transferability between the source and the target domains. Analyzing whether the benefits of transferred knowledge are equally or unequally accessible and applicable across different domains or tasks speaks to the requirement of symmetry or asymmetry in transfer learning. Moreover, these TL performance metrics inform on the possibility of the occurrence of negative transfers and also provide insight into the possible vulnerability of the network to catastrophic forgetting. Finally, we discuss a sensitivity analysis of an Ensemble TL technique use case (with Multilayer Perceptron models) as a proof of concept to validate the suggested framework. While the results from the experiments offer empirical insights into various parameters that impact the transfer learning gain, they also raise the question of network dimensioning requirements when designing, specifically, a neural network for transfer learning.

In this article, we focus on prediction of a target model by transferring the information of source models. To be flexible, we use semiparametric additive frameworks for the target and source models. Inheriting the spirit of parameter-transfer learning, we assume that different models possibly share common knowledge across parametric components that is helpful for the target predictive task. Unlike existing parameter-transfer approaches, which need to construct auxiliary source models by parameter similarity with the target model and then adopt a regularization procedure, we propose a frequentist model averaging strategy with a J-fold cross-validation criterion so that auxiliary parameter information from different models can be adaptively transferred through data-driven weight assignments. The asymptotic optimality and weight convergence of our proposed method are built under some regularity conditions. Extensive numerical results demonstrate the superiority of the proposed method over competitive methods. [ABSTRACT FROM AUTHOR]

This paper examines the use of prepositions in the L2 English speech production of L1 speakers of Czech. The data is sourced from a spoken corpus comprising forty c.15-minute interviews. L2 English preposition use was studied by means of potential occasion analysis, and the results show that prepositions pose a challenge to EFL learners. Careful scrutiny of the data revealed an increasing tendency toward preposition accuracy in speech across proficiency levels A2 to B2. Moreover, it is hypothesized that the participants' incorrect EFL preposition selection is influenced by their L1 knowledge. [ABSTRACT FROM AUTHOR]

This article demonstrates how Cognitive Grammar and Construction Grammar can prevent Chinese students learning Spanish from fossilizing mistakes in restrictive relative clauses at the A2-B1 level of the European Framework of Reference for Languages. To address this issue, first, relative clauses in Spanish and Chinese were contrasted and, second, tailored solutions based on Cognitive Grammar were proposed. Among the cognitive based tailored solutions, certain geometry forms, colours and basic mathematics metaphors were compared with syntactic characteristics such as noun order, subordination hierarchy or resumption. To elucidate the impact of such teaching methods, an experiment with 74 Chinese students was performed. The results indicate that the efficacy of the proposed materials is statistically significant and as such, the Chinese students avoid fossilized mistakes while producing subject, object and locative relative clauses in Spanish. [ABSTRACT FROM AUTHOR]

Copyright of Frontiers of Information Technology & Electronic Engineering is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Language teachers and researchers argue that the process of learning a language is better understood if the errors that language learners make in constructing the new language system are carefully analyzed. Hence, analyzing errors has become essential for facing and overcoming problems and suggesting solutions regarding different aspects of language teaching and learning. Accordingly, this study seeks to examine the errors that Moroccan high school students of English make in writing, and investigate the reasons behind the occurrence of such errors. The sample of the study consists of students' essays collected from different classes. The data collected were analyzed using Corder's (1967) model, which includes three stages; data collection, description, and explanation. The study's findings showed that Moroccan EFL students make different types of errors in writing. The most significant number of the errors found in the students' essays are grammatical errors, and most of these errors are due to intralingual factors. [ABSTRACT FROM AUTHOR]

In the context of data-driven deep learning, employing multi-fidelity methods for swift and precise wake field prediction is a novel attempt. Current Multi-Fidelity Neural Networks (MFNNs) face accuracy loss due to structure theory limitations, and performance degradation from negative transfer. To address these issues and enhance wake prediction performance, we propose a Dual-Adaptive Multi-Fidelity Neural Network (Ada2MF) framework. This framework features an adaptive multi-fidelity (AMF) module that integrates three sub-networks via a learnable weighted gate, effectively capturing the linear, nonlinear, and residual characteristics of high-fidelity data and mitigating theoretical accuracy loss. Additionally, the adaptive fast weighting (AFW) module employs a dynamic loss-weighting algorithm to optimally balance multi-fidelity losses and prevent negative transfer. Initial validation on benchmark functions and further evaluation using a multi-fidelity wind turbine wake field database confirm the effectiveness of Ada2MF. Specifically, with a complete dataset, Ada2MF achieves 66% and 41% improvements in wake prediction accuracy over single-fidelity neural networks and MFNN, respectively. Even with an 80% reduction in data volume, these improvements escalate to 82% and 78%, without incurring significant accuracy loss. Such results underscore Ada2MF's remarkable ability to improve prediction accuracy while substantially reducing the dependency on high-fidelity data. [Display omitted] [ABSTRACT FROM AUTHOR]

Universal domain adaptive object detection (UniDAOD) is a more challenging and realistic problem than traditional domain adaptive object detection (DAOD), aiming to transfer the knowledge from the well-labeled source domain to the unlabeled target domain without any prior knowledge of label sets. Intuitively, the main challenge of UniDAOD is to eliminate the domain shift and suppress the interference caused by the category shift induced by private classes (i.e., classes only existed in one domain). In the current study, we propose a simple but effective CODE framework, namely Co nfused and D isentangled E xtraction, for alleviating this issue. Specifically, we propose the virtual adversarial adaptation module, characterized by incorporating virtual domain labels within the domain classifier for unaligned samples. This confuses the domain classifier, effectively addressing the issue of converging to local optima resulting from equilibrium challenges and consequently narrowing the domain shift. Simultaneously, we introduce the entropy margin separation module, which utilizes the distinctiveness of category predictions as a disentangled factor. This enables the automatic discovery of private classes in each domain, suppressing interference during the adaptation process. Experiments on four universal scenarios (i.e., closed-set, partial-set, open-partial-set, and open-set) show that CODE obtains a significant performance gain over original DAOD detectors. [ABSTRACT FROM AUTHOR]

In the realm of object detection, traditional domain adaptive object detection (DAOD) methods assume that source and target data completely share one identical class space, which is often difficult to satisfy in many real-world applications. To address this limitation, this paper introduces universal domain adaptive object detection (UniDAOD), a learning paradigm that relaxes identical class space assumption to be a different but overlapped class space. Intuitively, the main challenge of UniDAOD is to reduce the negative transfer of private classes (i.e., classes only existed in one domain) and reinforce the positive transfer of the common classes (i.e., classes shared across domains). In this paper, we provide a rigorous theoretical analysis and induce a new generalization bound of the expected target error under the UniDAOD setting. On the basis of theoretical insight, we then propose weighted adaptation (W-adapt) to suppress the interference of private classes and reinforce the positive effects of common classes. In particular, we propose a pseudo category margin (PCM) to quantify class importance based on dynamic pseudotarget label prediction to recognize common classes. Furthermore, to alleviate the impact of inaccurate pseudotarget labels, we propose a temporary memory-based filter (TMF) to dynamically store and update the PCM during progressive training. On the basis of the learned TMF, we design a weighted classwise domain alignment loss to adapt two domains across common classes. Experiments on four universal scenarios (i.e., partial-set, open-partial-set, open-set, and closed-set) show that W-adapt outperforms several domain adaptation methods. [ABSTRACT FROM AUTHOR]

Transfer learning algorithm can provide a framework to utilize the previous knowledge to train fuzzy neural network (FNN). However, the performance of TL-based FNN will be destroyed by the knowledge over-fitting problem in the learning process. To solve this problem, an interactive transfer learning (ITL) algorithm, which can alleviate the negative transfer among different domains to improve the learning performance of FNN, is designed and analyzed in this article. This ITL-assisted FNN (ITL-FNN) contains the following advantages. First, a knowledge filter algorithm is developed to reconstruct the knowledge in source scene by balancing the matching accuracy and diversity. Then, the knowledge from source scene can fit the instance of target scene with suitable accuracy. Second, a self-balancing mechanism is designed to balance the driven information between the source and target scenes. Then, the knowledge can be refitted to reduce the useless information. Third, a structural competition algorithm is proposed to adjust the knowledge of FNN. Then, the proposed ITL-FNN can achieve compact structure to improve the generalization performance. Finally, some benchmark problems and industrial applications are provided to demonstrate the merits of ITL-FNN. [ABSTRACT FROM AUTHOR]