An efficient optimization based microstructure reconstruction approach with multiple loss functions

In this essay, we describe the construction and use of the Cut-Score Operating Function in aiding standard setting decisions. The Cut-Score Operating Function shows the relation between the cut-score chosen and the consequent error rate. It allows error rates to be defined by multiple loss functions and will show the behavior of each loss function. One strength of the Cut-Score Operating Function is that it shows how robust error rates are to the choice of cut-score and identifies the regions of extreme sensitivity relative to that choice.

Download Full-text

Interpretable Multiple Loss Functions in A Low-Rank Deep Image Prior Based Method For Single Hyperspectral Image Super-Resolution

10.23919/eusipco54536.2021.9616146 ◽

2021 ◽

Author(s):

Tatiana Gelvez ◽

Jorge Bacca ◽

Henry Arguello

Keyword(s):

Hyperspectral Image ◽

Super Resolution ◽

Loss Functions ◽

Low Rank ◽

Image Prior ◽

Deep Image ◽

Image Super Resolution ◽

Multiple Loss

Download Full-text

Valence electron spectroscopy of inhomogeneous media

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100130389 ◽

1992 ◽

Vol 50 (2) ◽

pp. 1150-1151

Author(s):

A. Howie ◽

D.W. McComb

Keyword(s):

Specific Gravity ◽

Loss Function ◽

Electron Spectroscopy ◽

Valence Electron ◽

Peak Height ◽

Loss Functions ◽

Loss Peak ◽

High Energies ◽

Valence Electron Density ◽

Electron Microscopist

The bulk loss function Im(-l/ε (ω)), a well established tool for the interpretation of valence loss spectra, is being progressively adapted to the wide variety of inhomogeneous samples of interest to the electron microscopist. Proportionality between n, the local valence electron density, and ε-1 (Sellmeyer's equation) has sometimes been assumed but may not be valid even in homogeneous samples. Figs. 1 and 2 show the experimentally measured bulk loss functions for three pure silicates of different specific gravity ρ - quartz (ρ = 2.66), coesite (ρ = 2.93) and a zeolite (ρ = 1.79). Clearly, despite the substantial differences in density, the shift of the prominent loss peak is very small and far less than that predicted by scaling e for quartz with Sellmeyer's equation or even the somewhat smaller shift given by the Clausius-Mossotti (CM) relation which assumes proportionality between n (or ρ in this case) and (ε - 1)/(ε + 2). Both theories overestimate the rise in the peak height for coesite and underestimate the increase at high energies.

Download Full-text

A transfer learning based microstructure reconstruction approach using microstructural correlation functions

10.26226/morressier.5f5f8e69aa777f8ba5bd604c ◽

2020 ◽

Author(s):

Ashwini Gupta

Keyword(s):

Transfer Learning ◽

Correlation Functions ◽

Microstructure Reconstruction

Download Full-text

The NALCN Channel Regulator UNC-80 Functions in a Subset of Interneurons To Regulate Caenorhabditis elegans Reversal Behavior

G3 Genes|Genome|Genetics ◽

10.1534/g3.119.400692 ◽

2019 ◽

Vol 10 (1) ◽

pp. 199-210 ◽

Cited By ~ 1

Author(s):

Chuanman Zhou ◽

Jintao Luo ◽

Xiaohui He ◽

Qian Zhou ◽

Yunxia He ◽

...

Keyword(s):

Plant Hormone ◽

Neuronal Excitability ◽

Resting Membrane Potential ◽

Loss Of Function ◽

Wild Type ◽

C Elegans ◽

Link Type ◽

Complex Functions ◽

And Function ◽

Multiple Loss

NALCN (Na+leak channel, non-selective) is a conserved, voltage-insensitive cation channel that regulates resting membrane potential and neuronal excitability. UNC79 and UNC80 are key regulators of the channel function. However, the behavioral effects of the channel complex are not entirely clear and the neurons in which the channel functions remain to be identified. In a forward genetic screen for C. elegans mutants with defective avoidance response to the plant hormone methyl salicylate (MeSa), we isolated multiple loss-of-function mutations in unc-80 and unc-79. C. elegans NALCN mutants exhibited similarly defective MeSa avoidance. Interestingly, NALCN, unc-80 and unc-79 mutants all showed wild type-like responses to other attractive or repelling odorants, suggesting that NALCN does not broadly affect odor detection or related forward and reversal behaviors. To understand in which neurons the channel functions, we determined the identities of a subset of unc-80-expressing neurons. We found that unc-79 and unc-80 are expressed and function in overlapping neurons, which verified previous assumptions. Neuron-specific transgene rescue and knockdown experiments suggest that the command interneurons AVA and AVE and the anterior guidepost neuron AVG can play a sufficient role in mediating unc-80 regulation of the MeSa avoidance. Though primarily based on genetic analyses, our results further imply that MeSa might activate NALCN by direct or indirect actions. Altogether, we provide an initial look into the key neurons in which the NALCN channel complex functions and identify a novel function of the channel in regulating C. elegans reversal behavior through command interneurons.

Download Full-text

A CLASS OF LORENZ CURVES BASED ON LINEAR EXPONENTIAL LOSS FUNCTIONS

Communication in Statistics- Theory and Methods ◽

10.1081/sta-120004190 ◽

2002 ◽

Vol 31 (6) ◽

pp. 925-942 ◽

Cited By ~ 6

Author(s):

José María Sarabia ◽

Marta Pascual

Keyword(s):

Loss Functions ◽

Lorenz Curves

Download Full-text

Study on Radar Echo-Filling in an Occlusion Area by a Deep Learning Algorithm

Remote Sensing ◽

10.3390/rs13091779 ◽

2021 ◽

Vol 13 (9) ◽

pp. 1779

Author(s):

Xiaoyan Yin ◽

Zhiqun Hu ◽

Jiafeng Zheng ◽

Boyong Li ◽

Yuanyuan Zuo

Keyword(s):

Deep Learning ◽

Loss Function ◽

Learning Algorithm ◽

Weather Radar ◽

Loss Functions ◽

Training Dataset ◽

Echo Intensity ◽

Common Mean ◽

Deep Learning Algorithm ◽

Radar Beam

Radar beam blockage is an important error source that affects the quality of weather radar data. An echo-filling network (EFnet) is proposed based on a deep learning algorithm to correct the echo intensity under the occlusion area in the Nanjing S-band new-generation weather radar (CINRAD/SA). The training dataset is constructed by the labels, which are the echo intensity at the 0.5° elevation in the unblocked area, and by the input features, which are the intensity in the cube including multiple elevations and gates corresponding to the location of bottom labels. Two loss functions are applied to compile the network: one is the common mean square error (MSE), and the other is a self-defined loss function that increases the weight of strong echoes. Considering that the radar beam broadens with distance and height, the 0.5° elevation scan is divided into six range bands every 25 km to train different models. The models are evaluated by three indicators: explained variance (EVar), mean absolute error (MAE), and correlation coefficient (CC). Two cases are demonstrated to compare the effect of the echo-filling model by different loss functions. The results suggest that EFnet can effectively correct the echo reflectivity and improve the data quality in the occlusion area, and there are better results for strong echoes when the self-defined loss function is used.

Download Full-text

Robustness of convolutional neural network models in hyperspectral noisy datasets with loss functions

Computers & Electrical Engineering ◽

10.1016/j.compeleceng.2021.107009 ◽

2021 ◽

Vol 90 ◽

pp. 107009

Author(s):

Sepehr Ghafari ◽

Milad Ghobadi Tarnik ◽

Hadi Sadoghi Yazdi

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Network Models ◽

Loss Functions ◽

Neural Network Models

Download Full-text

Testing Segmentation Popular Loss and Variations in Three Multiclass Medical Imaging Problems

Journal of Imaging ◽

10.3390/jimaging7020016 ◽

2021 ◽

Vol 7 (2) ◽

pp. 16

Author(s):

Pedro Furtado

Keyword(s):

False Negative ◽

Magnetic Resonance Images ◽

Medical Image Segmentation ◽

Cross Entropy ◽

Loss Functions ◽

Optic Disk ◽

False Negatives ◽

Convolutional Network ◽

Percentage Points ◽

Class Background

Image structures are segmented automatically using deep learning (DL) for analysis and processing. The three most popular base loss functions are cross entropy (crossE), intersect-over-the-union (IoU), and dice. Which should be used, is it useful to consider simple variations, such as modifying formula coefficients? How do characteristics of different image structures influence scores? Taking three different medical image segmentation problems (segmentation of organs in magnetic resonance images (MRI), liver in computer tomography images (CT) and diabetic retinopathy lesions in eye fundus images (EFI)), we quantify loss functions and variations, as well as segmentation scores of different targets. We first describe the limitations of metrics, since loss is a metric, then we describe and test alternatives. Experimentally, we observed that DeeplabV3 outperforms UNet and fully convolutional network (FCN) in all datasets. Dice scored 1 to 6 percentage points (pp) higher than cross entropy over all datasets, IoU improved 0 to 3 pp. Varying formula coefficients improved scores, but the best choices depend on the dataset: compared to crossE, different false positive vs. false negative weights improved MRI by 12 pp, and assigning zero weight to background improved EFI by 6 pp. Multiclass segmentation scored higher than n-uniclass segmentation in MRI by 8 pp. EFI lesions score low compared to more constant structures (e.g., optic disk or even organs), but loss modifications improve those scores significantly 6 to 9 pp. Our conclusions are that dice is best, it is worth assigning 0 weight to class background and to test different weights on false positives and false negatives.

Download Full-text