Deep Classification with Linearity-Enhanced Logits to Softmax Function

Abstract

Recently, there has been a rapid increase in deep classification tasks, such as image recognition and target detection. As one of the most crucial components in Convolutional Neural Network (CNN) architectures, softmax arguably encourages CNN to achieve better performance in image recognition. Under this scheme, we present a conceptually intuitive learning objection function: Orthogonal-Softmax. The primary property of the loss function is to use a linear approximation model that is designed by Gram–Schmidt orthogonalization. Firstly, compared with the traditional softmax and Taylor-Softmax, Orthogonal-Softmax has a stronger relationship through orthogonal polynomials expansion. Secondly, a new loss function is advanced to acquire highly discriminative features for classification tasks. At last, we present a linear softmax loss to further promote the intra-class compactness and inter-class discrepancy simultaneously. The results of the widespread experimental discussion on four benchmark datasets manifest the validity of the presented method. Besides, we want to explore the non-ground truth samples in the future.

Machine vision-based autonomous road hazard avoidance system for self-driving vehicles

“…Fully connected layers flatten output feature maps into one-dimensional vectors, which enter a network for classification. Finally, the SoftMax layer 21 transforms fully connected layer output into a probability distribution, crucial for classifying images in the vehicle vision system. Neural networks, with their multi-layered architecture and advanced feature extraction capabilities, enable thorough analysis of visual data from vehicles, providing essential environmental perception capabilities for autonomous driving systems.…”

Bridging the Diagnostic Gap between Histopathologic and Hysteroscopic Chronic Endometritis with Deep Learning Models

“…Throughout the fully connected layers, the feature maps are set up for the final outputs. The softmax function, defined as softmax(z j ) = exp(z i )/Σ exp(z i ), is an activation function that is typically applied to the multiclass classification task for the final outputs of the feature maps ( Figure 5 ) [ 79 ]. As the formula of the softmax function contains an exponential conversion of the outputs, the real output values become larger beyond a certain range of numbers.…”