Deep Domain-Adversarial Anomaly Detection With One-Class Transfer Learning

Despite the big success of transfer learning techniques in anomaly detection,it is still challenging to achieve good transition of detection rules merely based on the preferred data in the anomaly detection with one-class classification,especially for the data with a large distribution difference.To address this challenge,a novel deep one-class transfer learning algorithm with domain-adversarial training is proposed in this paper.First,by integrating a hypersphere adaptation constraint into domainadversarial neural network,a new hypersphere adversarial training mechanism is designed.Second,an alternative optimization method is derived to seek the optimal network parameters while pushing the hyperspheres built in the source domain and target domain to be as identical as possible.Through transferring oneclass detection rule in the adaptive extraction of domain-invariant feature representation,the end-to-end anomaly detection with one-class classification is then enhanced.Furthermore,a theoretical analysis about the model reliability,as well as the strategy of avoiding invalid and negative transfer,is provided.Experiments are conducted on two typical anomaly detection problems,i.e.,image recognition detection and online early fault detection of rolling bearings.The results demonstrate that the proposed algorithm outperforms the state-of-the-art methods in terms of detection accuracy and robustness.

Effect of microRNA-143-3p-mediated CTNND1 on the biological function of lung cancer cells

Lung cancer poses a serious threat to human life with high incidence and miRNA is an important biomarkerin tumors. This study aimed to explore the effect of miR-143-3p on the biological function of lung cancer cells and theunderlying mechanism. Eighty-seven samples of lung cancer tissues and 81 samples of tumor-adjacent tissues from patients undergoing radical lung cancer surgery in our hospital were collected. The lung cancer cells and lung fibroblastcells (HFL-1) were purchased, and then miR-143-3p-mimics, miR-NC, si-CTNND1, and NC were transfected into A549 and PC-9 cells to establish cell models. MiR-143-3p and CTNND1 expression levels were measured by the qRT-PCR, Bax, Bcl-2, and CTNND1 expression levels by the Western Blot (WB), and cell proliferation, invasion, and apoptosis by the MTT assay, Transwell assay, and flow cytometry. Dual luciferase report assay was used to determinethe relationship between miR-143-3p and CTNND1. In this study, miR-143-3p was lowly expressed in lung cancer and CTNND1 was highly expressed in lung cancer. The overexpression of miR-143-3p inhibited cell proliferation and invasion, promoted cell apoptosis, significantly increased Bax protein expression, and decreased Bcl-2 protein expression. The inhibition of CTNND1 led to opposite biological characteristic in cells. The dual luciferase reporter assay demonstrated that miR-143-3p was a target region of CTNND1. Such results suggest that miR-143-3p can inhibitthe proliferation and invasion of lung cancer cells by regulating the expression of CTNND1 and promote the apoptosisof lung cancer cells, sott is expected to be a potential target for lung cancer.

Water quality prediction based on sparse dataset using enhanced machine learning

Water quality in surface bodies remains a pressing issue worldwide.While some regions have rich water quality data,less attention is given to areas that lack sufficient data.Therefore,it is crucial to explore novel ways of managing source-oriented surface water pollution in scenarios with infrequent data collection such as weekly or monthly.Here we showed sparse-dataset-based prediction of water pollution using machine learning.We investigated the efficacy of a traditional Recurrent Neural Network alongside three Long Short-Term Memory(LSTM)models,integrated with the Load Estimator(LOADEST).The research was conducted at a river-lake confluence,an area with intricate hydrological patterns.We found that the Self-Attentive LSTM(SA-LSTM)model outperformed the other three machine learning models in predicting water quality,achieving Nash-Sutcliffe Efficiency(NSE)scores of 0.71 for COD_(Mn)and 0.57 for NH_(3)N when utilizing LOADEST-augmented water quality data(referred to as the SA-LSTMLOADEST model).The SA-LSTM-LOADEST model improved upon the standalone SA-LSTM model by reducing the Root Mean Square Error(RMSE)by 24.6%for COD_(Mn)and 21.3%for NH_(3)N.Furthermore,the model maintained its predictive accuracy when data collection intervals were extended from weekly to monthly.Additionally,the SA-LSTM-LOADEST model demonstrated the capability to forecast pollution loads up to ten days in advance.This study shows promise for improving water quality modeling in regions with limited monitoring capabilities.

Investigating drivers of microbial activity and respiration in a forested bog

Northern peatlands store nearly one-third of terrestrial carbon(C)stocks while covering only 3%of the global landmass;nevertheless,the drivers of C cycling in these often-waterlogged ecosystems are different from those that control C dynamics in upland forested soils.To explore how multiple abiotic and biotic characteristics of bogs interact to shape microbial activity in a northern,forested bog,we added a labile C tracer(13C-labeled starch)to in situ peat mesocosms and correlated heterotrophic respiration with natural variation in several microbial predictor variables,such as enzyme activity and microbial biomass,as well as with a suite of abiotic variables and proximity to vascular plants aboveground.We found that peat moisture content was positively correlated with respiration and microbial activity,even when moisture levels exceeded total saturation,suggesting that access to organic matter substrates in drier environments may be limiting for microbial activity.Proximity to black spruce trees decreased total and labile heterotrophic respiration.This negative relationship may reflect the influence of tree evapotranspiration and peat shading effects;i.e.,microbial activity may decline as peat dries and cools near trees.Here,we isolated the response of heterotrophic respiration to explore the variation in,and interactions among,multiple abiotic and biotic drivers that influence microbial activity.This approach allowed us to reveal the relative influence of individual drivers on C respiration in these globally important C sinks.