Search Results for author:
Found 21 papers, 15 papers with code
Towards Robust Deep Learning using Entropic Losses
This thesis tackles the defiant out-of-distribution detection task by proposing novel loss functions and detection scores.
Distinction Maximization Loss: Efficiently Improving Out-of-Distribution Detection and Uncertainty Estimation by Replacing the Loss and Calibrating
In this paper, we propose training deterministic neural networks using our DisMax loss, which works as a drop-in replacement for the usual SoftMax loss (i. e., the combination of the linear output layer, the SoftMax activation, and the cross-entropy loss).
Multi-Cue Adaptive Emotion Recognition Network
Expressing and identifying emotions through facial and physical expressions is a significant part of social interaction.
Enhanced Isotropy Maximization Loss: Seamless and High-Performance Out-of-Distribution Detection Simply Replacing the SoftMax Loss
The IsoMax loss works as a drop-in replacement of the SoftMax loss (i. e., the combination of the output linear layer, the SoftMax activation, and the cross-entropy loss) because swapping the SoftMax loss with the IsoMax loss requires no changes in the model's architecture or training procedures/hyperparameters.
Improving Entropic Out-of-Distribution Detection using Isometric Distances and the Minimum Distance Score
The entropic out-of-distribution detection solution comprises the IsoMax loss for training and the entropic score for out-of-distribution detection.
Training Aware Sigmoidal Optimizer
We compared the proposed approach with commonly used adaptive learning rate schedules such as Adam, RMSProp, and Adagrad.
Intrusion Detection for Cyber-Physical Systems using Generative Adversarial Networks in Fog Environment
In this paper, we propose FID-GAN, a novel fog-based, unsupervised intrusion detection system (IDS) for CPSs using GANs.
KutralNet: A Portable Deep Learning Model for Fire Recognition
The experiments show that our model keeps high accuracy while substantially reducing the number of parameters and flops.
Entropic Out-of-Distribution Detection: Seamless Detection of Unknown Examples
In this paper, we argue that the unsatisfactory out-of-distribution (OOD) detection performance of neural networks is mainly due to the SoftMax loss anisotropy and propensity to produce low entropy probability distributions in disagreement with the principle of maximum entropy.
Distantly-Supervised Neural Relation Extraction with Side Information using BERT
Considering that this method outperformed state-of-the-art baselines, in this paper, we propose a related approach to RESIDE also using additional side information, but simplifying the sentence encoding with BERT embeddings.
A Fast Fully Octave Convolutional Neural Network for Document Image Segmentation
In this context, we investigated a method based on U-Net to detect the document edges and text regions in ID images.
AM-MobileNet1D: A Portable Model for Speaker Recognition
To address this demand, we propose a portable model called Additive Margin MobileNet1D (AM-MobileNet1D) to Speaker Identification on mobile devices.
Squeezed Deep 6DoF Object Detection Using Knowledge Distillation
We used the LINEMOD dataset to evaluate the proposed method, and the experimental results show that the proposed method reduces the memory requirement by almost 99\% in comparison to the original architecture with the cost of reducing half the accuracy in one of the metrics.
Isotropy Maximization Loss and Entropic Score: Accurate, Fast, Efficient, Scalable, and Turnkey Neural Networks Out-of-Distribution Detection Based on The Principle of Maximum Entropy
Consequently, we propose IsoMax, a loss that is isotropic (distance-based) and produces high entropy (low confidence) posterior probability distributions despite still relying on cross-entropy minimization.
Spatial-Temporal Graph Convolutional Networks for Sign Language Recognition
The recognition of sign language is a challenging task with an important role in society to facilitate the communication of deaf persons.
Squeezed Very Deep Convolutional Neural Networks for Text Classification
Most of the research in convolutional neural networks has focused on increasing network depth to improve accuracy, resulting in a massive number of parameters which restricts the trained network to platforms with memory and processing constraints.
Ranked #17 on
Sentiment Analysis
on Yelp Fine-grained classification
Heartbeat Anomaly Detection using Adversarial Oversampling
As in most health problems, the imbalance between examples and classes is predominant in this problem and affects the performance of the automated solution.
Additive Margin SincNet for Speaker Recognition
The Softmax loss function is a widely used function in deep learning methods, but it is not the best choice for all kind of problems.
Hierarchical Attentional Hybrid Neural Networks for Document Classification
Document classification is a challenging task with important applications.
Ranked #1 on
Text Classification
on Yelp-5
Enhancing Batch Normalized Convolutional Networks using Displaced Rectifier Linear Units: A Systematic Comparative Study
Besides, statistical significant performance assessments (p<0. 05) showed DReLU enhanced the test accuracy presented by ReLU in all scenarios.
Simple Fast Convolutional Feature Learning
The quality of the features used in visual recognition is of fundamental importance for the overall system.
