scholarly journals Adversarially Robust Learning via Entropic Regularization

2022 ◽  
Vol 4 ◽  
Author(s):  
Gauri Jagatap ◽  
Ameya Joshi ◽  
Animesh Basak Chowdhury ◽  
Siddharth Garg ◽  
Chinmay Hegde
Keyword(s):  
Loss Function ◽  
Deep Neural Networks ◽  
Learning Approaches ◽  
High Loss ◽  
Robust Learning ◽  
Data Space ◽  
New Family ◽  
Training Samples ◽  
Benchmark Datasets ◽  
Entropic Regularization

In this paper we propose a new family of algorithms, ATENT, for training adversarially robust deep neural networks. We formulate a new loss function that is equipped with an additional entropic regularization. Our loss function considers the contribution of adversarial samples that are drawn from a specially designed distribution in the data space that assigns high probability to points with high loss and in the immediate neighborhood of training samples. Our proposed algorithms optimize this loss to seek adversarially robust valleys of the loss landscape. Our approach achieves competitive (or better) performance in terms of robust classification accuracy as compared to several state-of-the-art robust learning approaches on benchmark datasets such as MNIST and CIFAR-10.

scholarly journals An efficient pruning scheme of deep neural networks for Internet of Things applications

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Chen Qi ◽  
Shibo Shen ◽  
Rongpeng Li ◽  
Zhifeng Zhao ◽  
Qing Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Internet Of Things ◽  
Deep Neural Networks ◽  
Computational Cost ◽  
Superior Performance ◽  
Compact Structure ◽  
Resource Limited ◽  
Benchmark Datasets ◽  
Iot Devices

AbstractNowadays, deep neural networks (DNNs) have been rapidly deployed to realize a number of functionalities like sensing, imaging, classification, recognition, etc. However, the computational-intensive requirement of DNNs makes it difficult to be applicable for resource-limited Internet of Things (IoT) devices. In this paper, we propose a novel pruning-based paradigm that aims to reduce the computational cost of DNNs, by uncovering a more compact structure and learning the effective weights therein, on the basis of not compromising the expressive capability of DNNs. In particular, our algorithm can achieve efficient end-to-end training that transfers a redundant neural network to a compact one with a specifically targeted compression rate directly. We comprehensively evaluate our approach on various representative benchmark datasets and compared with typical advanced convolutional neural network (CNN) architectures. The experimental results verify the superior performance and robust effectiveness of our scheme. For example, when pruning VGG on CIFAR-10, our proposed scheme is able to significantly reduce its FLOPs (floating-point operations) and number of parameters with a proportion of 76.2% and 94.1%, respectively, while still maintaining a satisfactory accuracy. To sum up, our scheme could facilitate the integration of DNNs into the common machine-learning-based IoT framework and establish distributed training of neural networks in both cloud and edge.

scholarly journals Monocular Depth Estimation with Joint Attention Feature Distillation and Wavelet-Based Loss Function

Sensors ◽  
2020 ◽  
Vol 21 (1) ◽  
pp. 54
Author(s):  
Peng Liu ◽  
Zonghua Zhang ◽  
Zhaozong Meng ◽  
Nan Gao
Keyword(s):  
Joint Attention ◽  
Loss Function ◽  
Depth Estimation ◽  
Depth Information ◽  
3D Vision ◽  
Network Training ◽  
Crucial Component ◽  
Benchmark Datasets ◽  
Ill Posed ◽  
Monocular Depth

Depth estimation is a crucial component in many 3D vision applications. Monocular depth estimation is gaining increasing interest due to flexible use and extremely low system requirements, but inherently ill-posed and ambiguous characteristics still cause unsatisfactory estimation results. This paper proposes a new deep convolutional neural network for monocular depth estimation. The network applies joint attention feature distillation and wavelet-based loss function to recover the depth information of a scene. Two improvements were achieved, compared with previous methods. First, we combined feature distillation and joint attention mechanisms to boost feature modulation discrimination. The network extracts hierarchical features using a progressive feature distillation and refinement strategy and aggregates features using a joint attention operation. Second, we adopted a wavelet-based loss function for network training, which improves loss function effectiveness by obtaining more structural details. The experimental results on challenging indoor and outdoor benchmark datasets verified the proposed method’s superiority compared with current state-of-the-art methods.

scholarly journals Deep Learning of Appearance Affinity for Multi-Object Tracking and Re-Identification: A Comparative View

Electronics ◽  
2020 ◽  
Vol 9 (11) ◽  
pp. 1757
Author(s):  
María J. Gómez-Silva ◽  
Arturo de la Escalera ◽  
José M. Armingol
Keyword(s):  
Deep Learning ◽  
Object Tracking ◽  
Loss Function ◽  
Neural Model ◽  
Training Data ◽  
Learning Approaches ◽  
The Core ◽  
Triplet Loss ◽  
Affinity Model

Recognizing the identity of a query individual in a surveillance sequence is the core of Multi-Object Tracking (MOT) and Re-Identification (Re-Id) algorithms. Both tasks can be addressed by measuring the appearance affinity between people observations with a deep neural model. Nevertheless, the differences in their specifications and, consequently, in the characteristics and constraints of the available training data for each one of these tasks, arise from the necessity of employing different learning approaches to attain each one of them. This article offers a comparative view of the Double-Margin-Contrastive and the Triplet loss function, and analyzes the benefits and drawbacks of applying each one of them to learn an Appearance Affinity model for Tracking and Re-Identification. A batch of experiments have been conducted, and their results support the hypothesis concluded from the presented study: Triplet loss function is more effective than the Contrastive one when an Re-Id model is learnt, and, conversely, in the MOT domain, the Contrastive loss can better discriminate between pairs of images rendering the same person or not.

New active learning algorithms for near-infrared spectroscopy in agricultural applications

2021 ◽  
Vol 69 (4) ◽  
pp. 297-306
Author(s):  
Julius Krause ◽  
Maurice Günder ◽  
Daniel Schulz ◽  
Robin Gruna
Keyword(s):  
Active Learning ◽  
Near Infrared ◽  
Agricultural Products ◽  
Training Data ◽  
Calibration Model ◽  
Learning Approaches ◽  
Training Samples ◽  
Agricultural Applications ◽  
Selection Of

Abstract The selection of training data determines the quality of a chemometric calibration model. In order to cover the entire parameter space of known influencing parameters, an experimental design is usually created. Nevertheless, even with a carefully prepared Design of Experiment (DoE), redundant reference analyses are often performed during the analysis of agricultural products. Because the number of possible reference analyses is usually very limited, the presented active learning approaches are intended to provide a tool for better selection of training samples.

scholarly journals A Survey of Graphical Page Object Detection with Deep Neural Networks

2021 ◽  
Author(s):  
Jwalin Bhatt ◽  
Khurram Azeem Hashmi ◽  
Muhammad Zeshan Afzal ◽  
Didier Stricker
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Conceptual Understanding ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Learning Approaches ◽  
Document Images ◽  
Essential Information ◽  
Current State ◽  
High Level

In any document, graphical elements like tables, figures, and formulas contain essential information. The processing and interpretation of such information require specialized algorithms. Off-the-shelf OCR components cannot process this information reliably. Therefore, an essential step in document analysis pipelines is to detect these graphical components. It leads to a high-level conceptual understanding of the documents that makes digitization of documents viable. Since the advent of deep learning, the performance of deep learning-based object detection has improved many folds. In this work, we outline and summarize the deep learning approaches for detecting graphical page objects in the document images. Therefore, we discuss the most relevant deep learning-based approaches and state-of-the-art graphical page object detection in document images. This work provides a comprehensive understanding of the current state-of-the-art and related challenges. Furthermore, we discuss leading datasets along with the quantitative evaluation. Moreover, it discusses briefly the promising directions that can be utilized for further improvements.

scholarly journals Self-Amplificated Network: Learning fine-grained learner with few samples

2021 ◽  
Vol 2050 (1) ◽  
pp. 012006
Author(s):  
Xili Dai ◽  
Chunmei Ma ◽  
Jingwei Sun ◽  
Tao Zhang ◽  
Haigang Gong ◽  
...  
Keyword(s):  
Deep Neural Networks ◽  
Classification Problem ◽  
The Self ◽  
Superior Performance ◽  
Query Image ◽  
Network Learning ◽  
Fine Grained ◽  
Support Set ◽  
Meta Learning ◽  
Benchmark Datasets

Abstract Training deep neural networks from only a few examples has been an interesting topic that motivated few shot learning. In this paper, we study the fine-grained image classification problem in a challenging few-shot learning setting, and propose the Self-Amplificated Network (SAN), a method based on meta-learning to tackle this problem. The SAN model consists of three parts, which are the Encoder, Amplification and Similarity Modules. The Encoder Module encodes a fine-grained image input into a feature vector. The Amplification Module is used to amplify subtle differences between fine-grained images based on the self attention mechanism which is composed of multi-head attention. The Similarity Module measures how similar the query image and the support set are in order to determine the classification result. In-depth experiments on three benchmark datasets have showcased that our network achieves superior performance over the competing baselines.

scholarly journals Explicit Interaction Model towards Text Classification

2019 ◽  
Vol 33 ◽  
pp. 6359-6366 ◽  
Author(s):  
Cunxiao Du ◽  
Zhaozheng Chen ◽  
Fuli Feng ◽  
Lei Zhu ◽  
Tian Gan ◽  
...  
Keyword(s):  
Language Processing ◽  
Text Classification ◽  
Deep Neural Networks ◽  
Interaction Mechanism ◽  
Interaction Model ◽  
Classification Task ◽  
Fine Grained ◽  
Word Level ◽  
Benchmark Datasets ◽  
Classification Tasks

Text classification is one of the fundamental tasks in natural language processing. Recently, deep neural networks have achieved promising performance in the text classification task compared to shallow models. Despite of the significance of deep models, they ignore the fine-grained (matching signals between words and classes) classification clues since their classifications mainly rely on the text-level representations. To address this problem, we introduce the interaction mechanism to incorporate word-level matching signals into the text classification task. In particular, we design a novel framework, EXplicit interAction Model (dubbed as EXAM), equipped with the interaction mechanism. We justified the proposed approach on several benchmark datasets including both multilabel and multi-class text classification tasks. Extensive experimental results demonstrate the superiority of the proposed method. As a byproduct, we have released the codes and parameter settings to facilitate other researches.

scholarly journals Three‐dimensional dose prediction for lung IMRT patients with deep neural networks: robust learning from heterogeneous beam configurations

2019 ◽  
Vol 46 (8) ◽  
pp. 3679-3691 ◽  
Author(s):  
Ana María Barragán‐Montero ◽  
Dan Nguyen ◽  
Weiguo Lu ◽  
Mu-Han Lin ◽  
Roya Norouzi‐Kandalan ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Three Dimensional ◽  
Robust Learning ◽  
Dose Prediction

scholarly journals MSpectraAI: a powerful platform for deciphering proteome profiling of multi-tumor mass spectrometry data by using deep neural networks

2020 ◽  
Vol 21 (1) ◽  
Author(s):  
Shisheng Wang ◽  
Hongwen Zhu ◽  
Hu Zhou ◽  
Jingqiu Cheng ◽  
Hao Yang
Keyword(s):  
Mass Spectrometry ◽  
Neural Networks ◽  
Large Scale ◽  
Deep Neural Networks ◽  
Spectral Feature ◽  
Mass Spectrometry Data ◽  
Learning Approaches ◽  
Proteomics Data ◽  
Proteome Profiling ◽  
Analytical Technique

Abstract Background Mass spectrometry (MS) has become a promising analytical technique to acquire proteomics information for the characterization of biological samples. Nevertheless, most studies focus on the final proteins identified through a suite of algorithms by using partial MS spectra to compare with the sequence database, while the pattern recognition and classification of raw mass-spectrometric data remain unresolved. Results We developed an open-source and comprehensive platform, named MSpectraAI, for analyzing large-scale MS data through deep neural networks (DNNs); this system involves spectral-feature swath extraction, classification, and visualization. Moreover, this platform allows users to create their own DNN model by using Keras. To evaluate this tool, we collected the publicly available proteomics datasets of six tumor types (a total of 7,997,805 mass spectra) from the ProteomeXchange consortium and classified the samples based on the spectra profiling. The results suggest that MSpectraAI can distinguish different types of samples based on the fingerprint spectrum and achieve better prediction accuracy in MS1 level (average 0.967). Conclusion This study deciphers proteome profiling of raw mass spectrometry data and broadens the promising application of the classification and prediction of proteomics data from multi-tumor samples using deep learning methods. MSpectraAI also shows a better performance compared to the other classical machine learning approaches.

scholarly journals Combining Spectral Unmixing and 3D/2D Dense Networks with Early-Exiting Strategy for Hyperspectral Image Classification

2020 ◽  
Vol 12 (5) ◽  
pp. 779 ◽  
Author(s):  
Bei Fang ◽  
Yunpeng Bai ◽  
Ying Li
Keyword(s):  
Deep Learning ◽  
Network Performance ◽  
Hyperspectral Image ◽  
Spectral Unmixing ◽  
Classification Methods ◽  
Dense Network ◽  
Training Samples ◽  
Mixed Pixels ◽  
Benchmark Datasets ◽  
Hard Samples

Recently, Hyperspectral Image (HSI) classification methods based on deep learning models have shown encouraging performance. However, the limited numbers of training samples, as well as the mixed pixels due to low spatial resolution, have become major obstacles for HSI classification. To tackle these problems, we propose a resource-efficient HSI classification framework which introduces adaptive spectral unmixing into a 3D/2D dense network with early-exiting strategy. More specifically, on one hand, our framework uses a cascade of intermediate classifiers throughout the 3D/2D dense network that is trained end-to-end. The proposed 3D/2D dense network that integrates 3D convolutions with 2D convolutions is more capable of handling spectral-spatial features, while containing fewer parameters compared with the conventional 3D convolutions, and further boosts the network performance with limited training samples. On another hand, considering the existence of mixed pixels in HSI data, the pixels in HSI classification are divided into hard samples and easy samples. With the early-exiting strategy in these intermediate classifiers, the average accuracy can be improved by reducing the amount of computation cost for easy samples, thus focusing on classifying hard samples. Furthermore, for hard samples, an adaptive spectral unmixing method is proposed as a complementary source of information for classification, which brings considerable benefits to the final performance. Experimental results on four HSI benchmark datasets demonstrate that the proposed method can achieve better performance than state-of-the-art deep learning-based methods and other traditional HSI classification methods.

Sign in / Sign up

Export Citation Format

Share Document