Guided parallelized stochastic gradient descent for delay compensation

Sharma, Anuraganand

doi:10.1016/j.asoc.2021.107084

Cited by 16 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S j ( y i , x , i ) is defined as the state feature function corresponding to the marked position I in the observation sequence and it is applied to describe the influence of the observation sequence on the marked variable, as shown in Equation (4); λ j and μ k are parameters, and Z refers to the normalization factor to ensure the probability is correctly defined. The values of parameters λ j and μ k are estimated based on the stochastic gradient descent algorithm in this study because it has a faster convergence rate than other gradient descent algorithms (Sharma, 2021).

t_{j} ()(, y_{i + 1}, x, i) goodbreak= ({, \begin{array}{l} 1 if y_{i + 1} goodbreak= y_{a}, y_{i} goodbreak= y_{b} and x_{i} goodbreak= x_{a} \\ 0 otherwise \end{array})

s_{j} ()(, y_{i}, x, i) goodbreak= ({, \begin{array}{l} 1 if y_{i} goodbreak= y_{b} and x_{i} goodbreak= x_{a} \\ 0 otherwise \end{array})

The details for POI name extraction from Chinese texts based on the improved CRF algorithm in this study mainly include sequence annotation, feature selection, feature annotation, template design, and model training and testing.…”

Section: Methodsmentioning

confidence: 99%

Point‐of‐interest detection from Weibo data for map updating

Yang

Gao

Zheng

et al. 2022

Transactions in GIS

View full text Add to dashboard Cite

Points‐of‐interest (POIs) geographic information system data are increasingly important for supporting map generation and navigation services, although updating their semantic and location information still largely depends on manual labor. In this study, we propose a novel method to automatically detect the changes in POIs from Chinese text and check‐in position data provided by the Chinese social media platform, Weibo. The proposed method includes three steps: (1) POI name recognition; (2) location confirmation; (3) and change detection. First, we propose recognizing a POI's name from Weibo text using the improved conditional random field algorithm. Then, we detect the location of each named POI by integrating the text address with the check‐in position. The changes in the detected POIs are recognized by extracting the status words from Weibo text and a three‐level status word database. To verify the effectiveness of the proposed method, we examine Wuhan as a case and detect the changes in the commercial POI using real‐world Weibo data collected from January to September 2020. Based on the validation of three common map platforms, the data provided and the manual field investigation of 55 random samples, the identification accuracies for newly added POIs, the unchanged POIs, and expired POIs are approximately 100, 95.8, and 91.7%, respectively.

show abstract

t_{j} ()(, y_{i + 1}, x, i) goodbreak= ({, \begin{array}{l} 1 if y_{i + 1} goodbreak= y_{a}, y_{i} goodbreak= y_{b} and x_{i} goodbreak= x_{a} \\ 0 otherwise \end{array})

s_{j} ()(, y_{i}, x, i) goodbreak= ({, \begin{array}{l} 1 if y_{i} goodbreak= y_{b} and x_{i} goodbreak= x_{a} \\ 0 otherwise \end{array})

Section: Methodsmentioning

confidence: 99%

Point‐of‐interest detection from Weibo data for map updating

Yang

Gao

Zheng

et al. 2022

Transactions in GIS

View full text Add to dashboard Cite

show abstract

“…The time-complexity of GAN can be roughly given as O(nT Ld 2 ) where the new parameters L and T are layer-size and total iterations for a GAN. Its convergence rate with the Stochastic Gradient Descent would be O( 1T + σ 2 ) where σ 2 is the variance of the dataset [59]. See Section III for further details.…”

Section: B Generative Adversarial Network (Gan)mentioning

confidence: 99%

SMOTified-GAN for Class Imbalanced Pattern Classification Problems

2022

Self Cite

View full text Add to dashboard Cite

Class imbalance in a dataset is a major problem for classifiers that results in poor prediction with a high true positive rate (TPR) but a low true negative rate (TNR) for a majority positive training dataset. Generally, the pre-processing technique of oversampling of minority class(es) are used to overcome this deficiency. Our focus is on using the hybridization of Generative Adversarial Network (GAN) and Synthetic Minority Over-Sampling Technique (SMOTE) to address class imbalanced problems. We propose a novel two-phase oversampling approach involving knowledge transfer that has the synergy of SMOTE and GAN. The unrealistic or overgeneralized samples of SMOTE are transformed into realistic distribution of data by GAN where there is not enough minority class data available for GAN to process them by itself effectively. We named it SMOTified-GAN as GAN works on pre-sampled minority data produced by SMOTE rather than randomly generating the samples itself. The experimental results prove the sample quality of minority class(es) has been improved in a variety of tested benchmark datasets. Its performance is improved by up to 9% from the next best algorithm tested on F1-score measurements. Its time complexity is also reasonable which is around O(N 2 d 2 T ) for a sequential algorithm.INDEX TERMS Generative Adversarial Network (GAN), Synthetic Minority Over-Sampling Technique (SMOTE), SMOTified-GAN, class imbalance problem.

show abstract

“…A high

value has the risk of missing the lowest point of the slope. The GD algorithm performs two main steps: gradient computation and update of coefficients ( m , b ), which starts with the initialization of α, initial vector of [ m , b ] and the total number of iterations, then the algorithm stops when the termination criterion is meet, or just the overfitting starts [ 51 , 52 ]. In this paper, the GD algorithm stops when the shadowing achieves a mean value of 0 and random behavior.…”

Section: Post-processingmentioning

confidence: 99%

A Propagation Study of LoRa P2P Links for IoT Applications: The Case of Near-Surface Measurements over Semitropical Rivers

Gutiérrez-Gómez

Rangel

Edwards

et al. 2021

Sensors

View full text Add to dashboard Cite

Internet of Things (IoT) radio networks are becoming popular in several scenarios for short-range applications (e.g., wearables and home security) and medium-range applications (e.g., shipping container tracking and autonomous farming). They have also been proposed for water monitoring in flood warning systems. IoT communications may use long range (LoRa) radios working in the 915 MHz industrial, scientific and medical (ISM) band. In this research, we study the propagation characteristics of LoRa chirp radio signals close to and over water in a tropical meadow region. We use as a case study the Colima River in Mexico. We develop a novel point-to-point IoT measurement sounding system that does not require decoding of LoRa propriety bursts and provides accurate power versus distance profiles along the riparian zone of a steeply dropping mountain river. We used this system to obtain the measurements reported in this work, which are also analyzed and modeled. The results show that the LoRa signal propagation over water exhibits a log-normal distribution. As a result of the chirp signal processing, two new experimental path loss models are presented. The path loss results show a considerable degradation of the received signal power over water within vegetation and less signal degradation at antenna heights closer to the water surface.

show abstract

Guided parallelized stochastic gradient descent for delay compensation

Cited by 16 publications

References 15 publications

Point‐of‐interest detection from Weibo data for map updating

Point‐of‐interest detection from Weibo data for map updating

SMOTified-GAN for Class Imbalanced Pattern Classification Problems

A Propagation Study of LoRa P2P Links for IoT Applications: The Case of Near-Surface Measurements over Semitropical Rivers

Contact Info

Product

Resources

About