Auxiliary Domain Research Articles

Impact of dataset on the study of crop disease image recognition

Datasets are very important in image recognition research based on machine learning methods. In particular, advanced methods such as deep learning and transfer learning are more dependent on datasets used for training models. The quality of datasets directly affects the final effect of these methods. In the research of crop disease image recognition, due to the complication of the agricultural environment and the variety of crops, datasets are scarce at present. Therefore, more and more researches adopt methods based on transfer learning, which can make up for the lack of data in the target domain with the help of other datasets. Among these methods, the selection of auxiliary domain datasets has great impact on the modeling effect of target domain. In order to clarify the impact of datasets on the research of crop disease image recognition, this study used different deep neural network frameworks to study and compare the effects of different datasets in crop disease image recognition based on transfer learning. The selected datasets include PlantVillage and Image Database for Agricultural Diseases and Pests Research (IDADP), which have been widely used in recent studies. And the selected deep neural network frameworks include ResNet50, InceptionV3, and EfficientNet. In the method of this study, the datasets are preprocessed first, such as data enhancement. After dividing the auxiliary domain and the target domain, the selected deep neural network frameworks are used to pre-train the model on the auxiliary domain dataset. Finally, the parameter-based transfer learning method was used to construct the corresponding crop disease recognition model in the target. In the experiments, multiple different datasets and different models were tested and compared. The results show that when the test set samples and training sample scenarios are consistent, the recognition accuracy of different network frameworks on multiple test sets is generally high. When the scenarios of test set samples and training samples are inconsistent, the recognition of various test sets by different network models cannot obtain ideal results. For the recognition of crop disease images that are collected from the actual cultivation environment, the use of IDADP dataset modeling is better, and it has more practical value in the actual application of crop disease image recognition. Keywords: crop diseases, datasets, transfer learning, deep learning, image recognition DOI: 10.25165/j.ijabe.20221505.7005 Citation: Yuan Y, Chen L, Ren Y C, Wang S M, Li Y. Impact of dataset on the study for crop disease image recognition. Int J Agric & Biol Eng, 2022; 15(5): 181–186.

A New Adversarial Domain Generalization Network Based on Class Boundary Feature Detection for Bearing Fault Diagnosis

In recent years, many researchers have attempted to achieve cross-domain diagnosis of faults through domain adaptation (DA) methods. However, owing to the complex physical environments, applications of DA-based approach are not guaranteed to unknown operating environments. Some existing domain generalization (DG) methods require enough fully labeled source domains to train, which are often unavailable in practical settings. In this study, an adversarial domain generalization network (ADGN) based on class boundary feature detection is proposed. The ADGN can diagnose faults in unknown operating environments, and only one fully labeled domain is used in training. Although ADGN has to access fully unlabeled auxiliary domains, a large number of unlabeled samples exist under actual working conditions. In our method, fuzzy features at a classification boundary are detected by maximizing the classifier differences. Better feature mapping functions and domain-invariant features are obtained by adversarial training. As the training proceeds, the differences in the distribution of features among the source, auxiliary, and unknown domains become smaller so domain-invariant features can be used for fault diagnosis in unknown operating environments. Comprehensive experiments showed that ADGN can achieve higher fault diagnosis accuracies than other methods when only one fully labeled domain is used in an unknown operating environment. The ADGN can even cope comfortably with complex transfer tasks with different speeds and loads.

Discriminative Extreme Learning Machine with Cross‐Domain Mean Approximation for Unsupervised Domain Adaptation

Extreme Learning Machine (ELM) is widely used in various fields because of its fast training and high accuracy. However, it does not primarily work well for Domain Adaptation (DA) in which there are many annotated data from auxiliary domain and few even no annotated data in target domain. In this paper, we propose a new variant of ELM called Discriminative Extreme Learning Machine with Cross‐Domain Mean Approximation (DELM‐CDMA) for unsupervised domain adaptation. It introduces Cross‐Domain Mean Approximation (CDMA) into the hidden layer of ELM to reduce distribution discrepancy between domains for domain bias elimination, which is conducive to train a high accuracy ELM on annotated data from auxiliary domains for target tasks. Linear Discriminative Analysis (LDA) is also adopted to improve the discrimination of learned model and obtain higher accuracy. Moreover, we further provide a Discriminative Kernel Extreme Learning Machine with Cross‐Domain Mean Approximation (DKELM‐CDMA) as the kernelization extension of DELM‐CDMA. Some experiments are performed to investigate the proposed approach, and the result shows that DELM‐CDMA and DKELM‐CDMA could effectively extend ELM suitable for domain adaptation and outperform ELM and many other domain adaptation approaches.

The manifold embedded selective pseudo-labeling algorithm and transfer learning of small sample dataset

Special scene classification and identification tasks are not easily fulfilled to obtain samples, which results in a shortage of samples. The focus of current researches lies in how to use source domain data (or auxiliary domain data) to build domain adaption transfer learning models and to improve the classification accuracy and performance of small sample machine learning in these special and difficult scenes. In this paper, a model of deep convolution and Grassmann manifold embedded selective pseudo-labeling algorithm (DC-GMESPL) is proposed to enable transfer learning classifications among multiple small sample datasets. Firstly, DC-GMESPL algorithm uses satellite remote sensing image sample data as the source domain to extract the smoke features simultaneously from both the source domain and the target domain based on the Resnet50 deep transfer network. This is done for such special scene of the target domain as the lack of local sample data for forest fire smoke video images. Secondly, DC-GMESPL algorithm makes the source domain feature distribution aligned with the target domain feature distribution. The distance between the source domain and the target domain feature distribution is minimized by removing the correlation between the source domain features and re-correlation with the target domain. And then the target domain data is pseudo-labeled by selective pseudo-labeling algorithm in Grassmann manifold space. Finally, a trainable model is constructed to complete the transfer classification between small sample datasets. The model of this paper is evaluated by transfer learning between satellite remote sensing image and video image datasets. Experiments show that DC-GMESPL transfer accuracy is higher than DC-CMEDA, Easy TL, CMMS and SPL respectively. Compared with our former DC-CMEDA, the transfer accuracy of our new DC-GMESPL algorithm has been further improved. The transfer accuracy of DC-GMESPL from satellite remote sensing image to video image has been improved by 0.50%, the transfer accuracy from video image to satellite remote sensing image has been improved by 8.50% and then, the performance has been greatly improved.

A selective ensemble learning based two-sided cross-domain collaborative filtering algorithm

Recently, various Cross-Domain Collaborative Filtering (CDCF) algorithms are presented to address the sparsity problem, leveraging ratings of auxiliary domains to improve target domain’s recommendation performance. Therein, two-sided CDCF algorithms have shown better performance, given the fact that they can extract both user and item information. However, as the auxiliary domains are not all related to the target domain, utilizing information from all the auxiliary domains may not be optimal and would lead to low efficiency. A Two-Sided CDCF model based on Selective Ensemble learning considering both Accuracy and Efficiency (TSSEAE) is proposed to balance recommendation accuracy and efficiency. In TSSEAE, user-sided and item-sided auxiliary domains are firstly combined to improve performance of target domain. Then, CDCF problems are converted to ensemble learning problems, with each combination corresponding to a classifier. In this way, the problem of selecting combinations can be converted to that of selecting classifiers, which is a selective ensemble learning problem. Finally, a bi-objective optimization problem is solved to obtain Pareto optimal solutions for the selective ensemble learning problem. The experimental result on Amazon dataset shows the effectiveness of TSSEAE.

SRSF10: an atypical splicing regulator with critical roles in stress response, organ development, and viral replication

Serine/arginine splicing factor 10 (SRSF10) is a member of the family of mammalian splicing regulators known as SR proteins. Like several of its SR siblings, the SRSF10 protein is composed of an RNA binding domain (RRM) and of arginine and serine-rich auxiliary domains (RS) that guide interactions with other proteins. The phosphorylation status of SRSF10 is of paramount importance for its activity and is subjected to changes during mitosis, heat-shock, and DNA damage. SRSF10 overexpression has functional consequences in a growing list of cancers. By controlling the alternative splicing of specific transcripts, SRSF10 has also been implicated in glucose, fat, and cholesterol metabolism, in the development of the embryonic heart, and in neurological processes. SRSF10 is also important for the proper expression and processing of HIV-1 and other viral transcripts. We discuss how SRSF10 could become a potentially appealing therapeutic target to combat cancer and viral infections.

A mixed heterogeneous factorization model for non-overlapping cross-domain recommendation

Cross-domain collaborative filtering has attracted a great deal of attention for its capability of dealing with data sparsity by transferring valuable knowledge from auxiliary domains to assist in recommendation in a target domain. State-of-the-art research has mainly focused on the scenario in which auxiliary domains share users or items with a target domain. However, such auxiliary data is rare or the acquisition of them is limited due to privacy concerns in real-world applications. We investigate a more realistic scenario in which auxiliary domains have neither users nor items overlapped with a target domain, thereby facilitating the collection of auxiliary data. In order to extract and transfer the sharing knowledge between auxiliary domains and target domain, we assume the user-item feedback in each domain consist of two parts: domain-transferrable information containing the shared knowledge and domain-reserved information reflecting the domain-specific characteristics. Correspondingly, we propose a mixed heterogeneous factorization model to capture the sharing knowledge and the domain-specific characteristics based on adapted tensor factorization and biased matrix factorization respectively and then combine them together in an accumulative way. Meanwhile, three types of domain heterogeneity including preference heterogeneity, characteristics heterogeneity and rating bias are taken into account in this model. Experimenting on four publicly available datasets across different domains, we show that our model is superior to state-of-the-art methods in rating prediction and top-N recommendation tasks.

Molecular basis of mRNA transport by a kinesin-1-atypical tropomyosin complex.

Kinesin-1 carries cargos including proteins, RNAs, vesicles, and pathogens over long distances within cells. The mechanochemical cycle of kinesins is well described, but how they establish cargo specificity is not fully understood. Transport of oskar mRNA to the posterior pole of the Drosophila oocyte is mediated by Drosophila kinesin-1, also called kinesin heavy chain (Khc), and a putative cargo adaptor, the atypical tropomyosin, aTm1. How the proteins cooperate in mRNA transport is unknown. Here, we present the high-resolution crystal structure of a Khc-aTm1 complex. The proteins form a tripartite coiled coil comprising two in-register Khc chains and one aTm1 chain, in antiparallel orientation. We show that aTm1 binds to an evolutionarily conserved cargo binding site on Khc, and mutational analysis confirms the importance of this interaction for mRNA transport in vivo. Furthermore, we demonstrate that Khc binds RNA directly and that it does so via its alternative cargo binding domain, which forms a positively charged joint surface with aTm1, as well as through its adjacent auxiliary microtubule binding domain. Finally, we show that aTm1 plays a stabilizing role in the interaction of Khc with RNA, which distinguishes aTm1 from classical motor adaptors.

Distant Transfer Learning via Deep Random Walk

Transfer learning, which is to improve the learning performance in the target domain by leveraging useful knowledge from the source domain, often requires that those two domains are very close, which limits its application scope. Recently, distant transfer learning has been studied to transfer knowledge between two distant or even totally unrelated domains via unlabeled auxiliary domains that act as a bridge in the spirit of human transitive inference that two completely unrelated concepts can be connected through gradual knowledge transfer. In this paper, we study distant transfer learning by proposing a DeEp Random Walk basEd distaNt Transfer (DERWENT) method. Different from existing distant transfer learning models that implicitly identify the path of knowledge transfer between the source and target instances through auxiliary instances, the proposed DERWENT model can explicitly learn such paths via the deep random walk technique. Specifically, based on sequences identified by the random walk technique on a data graph where source and target data have no direct connection, the proposed DERWENT model enforces adjacent data points in a sequence to be similar, makes the ending data point be represented by other data points in the same sequence, and considers weighted classification losses of source data. Empirical studies on several benchmark datasets demonstrate that the proposed DERWENT algorithm yields the state-of-the-art performance.

ML-SIM: universal reconstruction of structured illumination microscopy images using transfer learning

Structured illumination microscopy (SIM) has become an important technique for optical super-resolution imaging because it allows a doubling of image resolution at speeds compatible with live-cell imaging. However, the reconstruction of SIM images is often slow, prone to artefacts, and requires multiple parameter adjustments to reflect different hardware or experimental conditions. Here, we introduce a versatile reconstruction method, ML-SIM, which makes use of transfer learning to obtain a parameter-free model that generalises beyond the task of reconstructing data recorded by a specific imaging system for a specific sample type. We demonstrate the generality of the model and the high quality of the obtained reconstructions by application of ML-SIM on raw data obtained for multiple sample types acquired on distinct SIM microscopes. ML-SIM is an end-to-end deep residual neural network that is trained on an auxiliary domain consisting of simulated images, but is transferable to the target task of reconstructing experimental SIM images. By generating the training data to reflect challenging imaging conditions encountered in real systems, ML-SIM becomes robust to noise and irregularities in the illumination patterns of the raw SIM input frames. Since ML-SIM does not require the acquisition of experimental training data, the method can be efficiently adapted to any specific experimental SIM implementation. We compare the reconstruction quality enabled by ML-SIM with current state-of-the-art SIM reconstruction methods and demonstrate advantages in terms of generality and robustness to noise for both simulated and experimental inputs, thus making ML-SIM a useful alternative to traditional methods for challenging imaging conditions. Additionally, reconstruction of a SIM stack is accomplished in less than 200 ms on a modern graphics processing unit, enabling future applications for real-time imaging. Source code and ready-to-use software for the method are available at http://ML-SIM.github.io.

A Privacy-Preserving Cross-Domain Healthcare Wearables Recommendation Algorithm Based on Domain-Dependent and Domain-Independent Feature Fusion.

Recently, recommender systems are applied to provide personalized recomendation for healthcare wearables. However, due to the sparsity problem, traditional recommendation algorithms are difficult to achieve desired performance. Considering that consumers often buy and rate other types of items on E-commerce platforms, we can leverage significant information in the auxiliary domains to improve the recommendation performance of healthcare wearables, which can be regarded as cross-domain recommendation. However, traditional cross-domain recommendation model cannot fully represent user's characteristics and fail to consider the leaks of original auxiliary domain ratings during the information transfer process. To overcome the two shortcomings, this paper proposes a Privacy-Preserving Cross-Domain Healthcare Wearables Recommendation algorithm (PPCDHWRec). Firstly, user's characteristics are divided into domain-dependent features and domain-independent features, which complement each other and fully depict the user's characteristics. Secondly, inspired by the latent factor model, we factorize the original rating information of each auxiliary domain by Funk-SVD and Orthogonal Nonnegative Matrix Tri-Factorization (ONMTF) model, to obtain user's domain-dependent and domain-independent features, respectively. Finally, the Factorization Machine algorithm is used to fuse the obtained user's features with the target domain information to provide the recommendation results. By hiding the item latent factors obtained in the factorization process, PPCDHWRec ensures that the original information cannot be inferred from the transferred user hidden vector. Hence, PPCDHWRec is a privacy-preserving recommendation model. Experiments on two groups of auxiliary domains, having high and low correlations with target domain, show the effectiveness of PPCDHWRec.

Multistage transfer learning technique for classifying rare medical datasets

It is said that about 8% of the people across the world are impacted by different kinds of rare diseases. Identifying such rare diseases accurately is a challenging task, as they exhibit common symptoms that may be incorrectly recognized as a common disease. As a result, the treatment is insufficient and by the time a diagnosis is made, it may be too late for the patient to survive. Therefore, any attempt towards early diagnosis of rare diseases becomes the need of the hour and several researchers are concentrating on machine learning techniques to do the same. This work proposes one such approach to diagnose rare disorders. A multi-level transfer learning (MLTL) framework with three models is designed to detect and classify rare diseases with very limited datasets, using the knowledge acquired from easily available datasets. The first model, for the source domain, uses the abundantly available non-medical images and learns the generalized features. The acquired knowledge is then transferred to the second model, for the intermediate and auxiliary domain, which is again commonly available and related to the target domain, and helps in learning the auxiliary or intermediate task. This information is then used to classify the final target domain, which consists of medical datasets that are very scarce. Experimental results with non-medical and the auxiliary CT abdomen images show good classification accuracies for identifying rare diseases related to the liver and kidneys. An area under receiver operating characteristic (ROC) curve of 0.90 and 0.89 are achieved for two different rare diseases, with just 2.08% of the source domain dataset, 6.6% of the intermediate domain dataset and less than 10% of the rare target domain dataset, when compared to the work reported in the literature.

Cross-domain recommendation based on latent factor alignment

Recently, various cross-domain recommendation (CDR) models are proposed to overcome the sparsity problem, which leverage relatively abundant rating data from the auxiliary domain to improve recommendation performance of target domain. Though matrix factorization-based collaborative filtering algorithms gain dominance in single-domain recommendation systems, they cannot be used directly in cross-domain cases as the obtained latent factors of the target and auxiliary domains may not be aligned, which will lead to inaccurate knowledge transfer from the auxiliary domain to the target one. A CDR model named CDCFLFA is presented in this paper to solve this problem. In CDCFLFA, firstly latent factors between the two domains are aligned based on pattern matching. Then, user preferences of the auxiliary domain are transferred to update the original user latent vectors of target domain. Finally, a linear least square problem is solved to compute the item latent vectors of target domain and thus unknown ratings are obtained according to the updated user and item latent vectors. CDCFLFA does not require the same user or item sets between the two domains. Extensive experiments are conducted, and the results show that CDCFLFA achieves smaller MAE and RMSE values and larger precision and recall than the previous single- and cross-domain recommendation methods. Hence, CDCFLFA can be regarded as an effective cross-domain extension of single-domain matrix factorization algorithm.

Auxiliary Domain Selection in Cross-Domain Collaborative Filtering

The problem of data sparsity largely limits the accuracy of recommender systems in collaborative filtering model. To alleviate the problem, cross-domain collaborative filteri ng was proposed by harnessing the information from the auxiliary domains. Previous works mainly focused on improving the model of utilizing the auxiliary information yet little on the selection of auxiliary domains, although it is observed that the result of recommendation depends on the characteristics of auxiliary dataset . In this paper, we study the validity of cross-domain collaborative filtering by movie recommendation via different auxiliary domains of different movie genres. Through extensive experiments we find that the numbe r of overlapping users between target domain and auxiliary domain is an indicator of choosing beneficial domains, while the low Ku llback-Leibler divergence between non-overlapping user ratings, rather than the overlapping user ratings, is much more significant. The results are helpful in selection of auxiliary domains in cross-domain collaborative filtering.

Genome-wide identification and characterization of DEAD-box helicase family associated with early somatic embryogenesis in Dimocarpus longan Lour

DEAD-box (DDX) proteins belong to the largest subfamily of RNA helicase SF2, which contributes to all biological processes of RNA metabolism in the plant kingdom. Till now, no significant data are available regarding studies on DDX in Somatic Embryogenesis (SE) of woody plants. It is important to investigate the biological function of the DlDDX family in longan SE. Thus, a comprehensive analysis of 58 longan DEAD-box (DlDDX) genes characterization was performed by genome-wide identification and transcript abundance validation analysis. Homologous evolution has revealed that some DlDDXs in longan had high sequence similarity with Mus musculus, Citrus and Saccharomyces cerevisiae, indicating that DlDDXs were highly conservative in the animal, plant, and microorganism. Remarkably, gene duplication, purifying selection, and alternative splicing events, and new auxiliary domains have likely contributed to the functional evolution of DlDDX, indicating that DlDDX appeared neofunctionalization in longan. Besides, DlDDX3, 15, 28, 36 might interact with protein complex (MAC3A, MAC3B, CDC5, CBP20) of miRNA biosynthesis. Notably, DlDDX28 contained a novel auxiliary domain (CAF-1 p150), which might contribute to DNA demethylation in longan early SE. 4 DlDDX genes significantly expressed not only in early SE and zygotic embryogenesis (ZE) but also up-regulated at high levels in ‘Honghezi’ and ‘Quanlongbaihe’ with abortive seeds, which are of great significance. Moreover, some DlDDXs presented abiotic stress-response dynamic expression patterns by ABA, SA, JA, and NaCl treatments during early SE. Hence, DEAD-box is essential to SE development and seed abortive in longan.

An enhanced dynamic interaction network for claim verification

Claim verification aims to verify the truthfulness of claims according to the evidence. In the real-world scenarios, the claim verification methods are required 1) to identify and filter out the noise; 2) to enable the interaction among evidence; and 3) to predict the labels in mixed label sets. However, the prior studies adopt either the implicit or cumbersome interaction processes, which disable them from capturing the commonalities among evidence nor reducing the impact of noise, and few of them consider the label prediction in mixed label sets. To alleviate these issues, we propose an Enhanced Dynamic Interaction Network (EDIN). The EDIN contains two main modules: A Gate-based Dynamic Interaction Module to promote the interaction of evidence and summarize the commonality to verify the claim, which employs hub node and gates to control the information transmission and denoise. An Enhancement Module to assist the EDIN with predicting labels in a mixed label sets environment, where we introduce the auxiliary domain detection task and label transition component to enhance the expressiveness of label representation. Experimental results on multiple public datasets reveal the superiority of our method, and further analysis validates the effectiveness of our proposed modules.

A CCA-Based Item-Side Alignment Method for Cross-Domain Recommendation System

For cross-domain recommendation, it can be divided into strong correlation and weak correlation problems according to the consistency between auxiliary domain and target domain. The weak correlation problem is more practical than the strong correlation problem, and the solution is more difficult. The difficulty lies in how to establish an effective transfer model, to make sure the auxiliary domain and the target domain can perform effective collaborative training. For weak correlation problem, if the item-side of auxiliary domain and the target domain are not aligned, or the transfer model has a strong dependency on the user-side of the auxiliary domain, it will seriously affect the effect of cross-domain recommendation. To solve these problems mentioned above, we propose a CCA-based item-side alignment method (CIAM) by introducing: (1) item side alignment method. We use CCA to align the item side between auxiliary and target domain, to intensify the weak correlation between 2 domains. (2) the transfer model of retaining the user feature of target domain. The CIAM retained user features of target domain in UV decomposition, that makes the transfer model could not destroy the user feature between 2 domains. The proposed CIAM can improve the assistance of auxiliary domain, and can avoid the influence of the needless user-side of the auxiliary do-main on cross-domain recommendation. By experimental analysis, it can be verified that the proposed CIAM algorithm has a better performance than general cross-domain recommendation methods.

Selective Knowledge Transfer for Cross-Domain Collaborative Recommendation

Data sparsity is a major challenge for collaborative filtering recommender systems. A promising solution is to utilize feedback or ratings from multiple domains to improve the performance of recommendations in a collective way, known as the cross-domain recommendation. Cross-domain recommendation using heterogeneous feedback is a popular solution, which transfers knowledge from the more easily available auxiliary binary feedback to improve the prediction performance of the target domain. Most of the existing work focuses on the transfer of knowledge between different domains from the same website, where user behavior data in different domains can be fully shared. The existing work mainly assumes that data from different domains can be fully shared. However, due to the constraints of business privacy policies, it is difficult to directly share exactly the same user behavior data between different e-commerce websites. It results in that the user’s latent factors learned in the auxiliary domain cannot be directly transferred to the target domain, otherwise, it will cause a negative transfer issue. In this article, we consider that the auxiliary domain with numerical ratings and target domains with binary feedbacks only share overlapping items rather than users. We propose a Selective Knowledge Transfer for Cross-domain Collaborative Recommendation, called SKT. The proposed SKT framework not only transfers the item’s latent factors learned from the auxiliary domain to the target domain, but also selectively transfers the user’s latent factors learned from the auxiliary domain to the target domain. In addition, due to the introduction of co-graph regularization of user graphs and item graphs, SKT can maintain respective intrinsic geometric structure within each domain and thus avoid negative transfer issue. Extensive experiments conducted on two real-world datasets show that our SKT method is significantly better than all baseline methods at various density levels.

Utterance Verification-Based Dysarthric Speech Intelligibility Assessment Using Phonetic Posterior Features

In the literature, the task of dysarthric speech intelligibility assessment has been approached through development of different low-level feature representations, subspace modeling, phone confidence estimation or measurement of automatic speech recognition system accuracy. This paper proposes a novel approach where the intelligibility is estimated as the percentage of correct words uttered by a speaker with dysarthria by matching and verifying utterances of the speaker with dysarthria against control speakers’ utterances in phone posterior feature space and broad phonetic posterior feature space. Experimental validation of the proposed approach on the UA-Speech database, with posterior feature estimators trained on the data from auxiliary domain and language, obtained a best Pearson's correlation coefficient ( $r$ ) of 0.950 and Spearman's correlation coefficient ( $\rho$ ) of 0.957. Furthermore, replacing control speakers’ speech with speech synthesized by a neural text-to-speech system obtained a best $r$ of 0.931 and $\rho$ of 0.961.

A Sparse Deep Transfer Learning Model and Its Application for Smart Agriculture

The introduction of deep transfer learning (DTL) further reduces the requirement of data and expert knowledge in various uses of applications, helping DNN‐based models effectively reuse information. However, it often transfers all parameters from the source network that might be useful to the task. The redundant trainable parameters restrict DTL in low‐computing‐power devices and edge computing, while small effective networks with fewer parameters have difficulty transferring knowledge due to structural differences in design. For the challenge of how to transfer a simplified model from a complex network, in this paper, an algorithm is proposed to realize a sparse DTL, which only transfers and retains the most necessary structure to reduce the parameters of the final model. Sparse transfer hypothesis is introduced, in which a compressing strategy is designed to construct deep sparse networks that distill useful information in the auxiliary domain, improving the transfer efficiency. The proposed method is evaluated on representative datasets and applied for smart agriculture to train deep identification models that can effectively detect new pests using few data samples.