检索结果-内蒙古大学图书馆

Explaining Representation Learning with Perceptual Components

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yarici, Yavuz Kokilepersaud, Kiran Prabhushankar, Mohit AlRegib, Ghassan OLIVES The Center for Signal and Information Processing CSIP School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States

Self-supervised models create representation spaces that lack clear semantic meaning. This interpretability problem of representations makes traditional explainability methods ineffective in this context. In this paper, we introduce a novel method to analyze representation spaces using three key perceptual components: color, shape, and texture. We employ selective masking of these components to observe changes in representations, resulting in distinct importance maps for each. In scenarios, where labels are absent, these importance maps provide more intuitive explanations as they are integral to the human visual system. Our approach enhances the interpretability of the representation space, offering explanations that resonate with human visual perception. We analyze how different training objectives create distinct representation spaces using perceptual components. Additionally, we examine the representation of images across diverse image domains, providing insights into the role of these components in different contexts. © 2024, CC BY.

关键词： Textures

Taxes Are All You Need: Integration of Taxonomical Hierarchy Relationships into the Contrastive Loss

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kokilepersaud, Kiran Yarici, Yavuz Prabhushankar, Mohit AlRegib, Ghassan OLIVES The Center for Signal and Information Processing CSIP School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States

In this work, we propose a novel supervised contrastive loss that enables the integration of taxonomic hierarchy information during the representation learning process. A supervised contrastive loss operates by enforcing that images with the same class label (positive samples) project closer to each other than images with differing class labels (negative samples). The advantage of this approach is that it directly penalizes the structure of the representation space itself. This enables greater flexibility with respect to encoding semantic concepts. However, the standard supervised contrastive loss only enforces semantic structure based on the downstream task (i.e. the class label). In reality, the class label is only one level of a hierarchy of different semantic relationships known as a taxonomy. For example, the class label is oftentimes the species of an animal, but between different classes there are higher order relationships such as all animals with wings being "birds". We show that by explicitly accounting for these relationships with a weighting penalty in the contrastive loss we can out-perform the supervised contrastive loss. Additionally, we demonstrate the adaptability of the notion of a taxonomy by integrating our loss into medical and noise-based settings that show performance improvements by as much as 7%. © 2024, CC BY.

关键词： Taxonomies

INTELLIGENT MULTI-VIEW TEST TIME AUGMENTATION

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ozturk, Efe Prabhushankar, Mohit AlRegib, Ghassan OLIVES The Center for Signal and Information Processing CSIP School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States

In this study, we introduce an intelligent Test Time Augmentation (TTA) algorithm designed to enhance the robustness and accuracy of image classification models against viewpoint variations. Unlike traditional TTA methods that indiscriminately apply augmentations, our approach intelligently selects optimal augmentations based on predictive uncertainty metrics. This selection is achieved via a two-stage process: the first stage identifies the optimal augmentation for each class by evaluating uncertainty levels, while the second stage implements an uncertainty threshold to determine when applying TTA would be advantageous. This methodological advancement ensures that augmentations contribute to classification more effectively than a uniform application across the dataset. Experimental validation across several datasets and neural network architectures validates our approach, yielding an average accuracy improvement of 1.73% over methods that use single-view images. This research underscores the potential of adaptive, uncertainty-aware TTA in improving the robustness of image classification in the presence of viewpoint variations, paving the way for further exploration into intelligent augmentation strategies. The code is available at: https://***/olivesgatech/Intelligent-Multi-View-TTA © 2024, CC BY.

关键词： Deep neural networks

HEX: Hierarchical Emergence Exploitation in Self-Supervised Algorithms

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kokilepersaud, Kiran Kim, Seulgi Prabhushankar, Mohit AlRegib, Ghassan OLIVES The Center for Signal and Information Processing CSIP School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA United States

In this paper, we propose an algorithm that can be used on top of a wide variety of self-supervised (SSL) approaches to take advantage of hierarchical structures that emerge during training. SSL approaches typically work through some invariance term to ensure consistency between similar samples and a regularization term to prevent global dimensional collapse. Dimensional collapse refers to data representations spanning a lower-dimensional subspace. Recent work has demonstrated that the representation space of these algorithms gradually reflects a semantic hierarchical structure as training progresses. Data samples of the same hierarchical grouping tend to exhibit greater dimensional collapse locally compared to the dataset as a whole due to sharing features in common with each other. Ideally, SSL algorithms would take advantage of this hierarchical emergence to have an additional regularization term to account for this local dimensional collapse effect. However, the construction of existing SSL algorithms does not account for this property. To address this, we propose an adaptive algorithm that performs a weighted decomposition of the denominator of the InfoNCE loss into two terms: local hierarchical and global collapse regularization respectively. This decomposition is based on an adaptive threshold that gradually lowers to reflect the emerging hierarchical structure of the representation space throughout training. It is based on an analysis of the cosine similarity distribution of samples in a batch. We demonstrate that this hierarchical emergence exploitation (HEX) approach can be integrated across a wide variety of SSL algorithms. Empirically, we show performance improvements of up to 5.6% relative improvement over baseline SSL approaches on classification accuracy on imagenet with 100 epochs of training. © 2024, CC BY.

关键词： Consensus algorithm

Semantically interpretable and controllable filter sets

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Prabhushankar, Mohit Kwon, Gukyeong Temel, Dogancan AlRegib, Ghassan Center for Signal and Information Processing School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332-0250 United States

In this paper, we generate and control semantically interpretable filters that are directly learned from natural images in an unsupervised fashion. Each semantic filter learns a visually interpretable local structure in conjunction with other filters. The significance of learning these interpretable filter sets is demonstrated on two contrasting applications. The first application is image recognition under progressive decolorization, in which recognition algorithms should be color-insensitive to achieve a robust performance. The second application is image quality assessment where objective methods should be sensitive to color degradations. In the proposed work, the sensitivity and lack thereof are controlled by weighing the semantic filters based on the local structures they represent. To validate the proposed approach, we utilize the CURE-TSR dataset for image recognition and the TID 2013 dataset for image quality assessment. We show that the proposed semantic filter set achieves state-of-The-Art performances in both datasets while maintaining its robustness across progressive distortions. Copyright © 2019, The Authors. All rights reserved.

关键词： Semantics

Convex cone volume analysis for finding endmembers in hyperspectral imagery

学校读者我要写书评

暂无评论

Convex cone volume analysis for finding endmembers in hypers...

作者： Chang, Chein-I Xiong, Wei Chen, Shih-Yu Information and Technology College Dalian Maritime University Dalian China School of Physics and Optoelectronic Engineering Xidian University Xian China Remote Sensing Signal and Image Processing Laboratory Department of Computer Science and Electrical Engineering University of Maryland Baltimore County BaltimoreMD21250 United States Department of Computer Science and Information Management Providence University Taichung Taiwan Department of Computer Science and Information Engineering National Yunlin University of Science and Technology Yunlin Taiwan

This paper presents a new approach, called convex cone volume analysis (CCVA), which can be considered as a partially constrained-abundance (abundance non-negativity constraint) technique to find endmembers. It can be shown that finding the maximal volume of a convex cone in the original data space is equivalent to finding the maximal volume of a simplex in a hyperplane. As a result, the CCVA can take advantage of many recently developed fast computational algorithms developed for N-FINDR to derive their counterparts for CCVA. © Copyright 2016 Inderscience Enterprises Ltd.

关键词： Spectroscopy

Implicit background estimation for semantic segmentation

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Lehman, Charles Temel, Dogancan AlRegib, Ghassan Center for Signal and Information Processing School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332-0250 United States

Scene understanding and semantic segmentation are at the core of many computer vision tasks, many of which, involve interacting with humans in potentially dangerous ways. It is therefore paramount that techniques for principled design of robust models be developed. In this paper, we provide analytic and empirical evidence that correcting potentially errant non-distinct mappings that result from the softmax function can result in improving robustness characteristics on a state-of-the-art semantic segmentation model with minimal impact to performance and minimal changes to the code base. Copyright © 2019, The Authors. All rights reserved.

关键词： Semantics

Single image Super-Resolution Based on Support Vector Regression

学校读者我要写书评

暂无评论

Single Image Super-Resolution Based on Support Vector Regres...

International Joint Conference on Neural Networks (IJCNN)

作者： Dalong Li Steven Simske Russell M. Mersereau Digital Printing and Imaging Lab Hewlett Packard Laboratories Fort Collins CO USA Center for Signal and Image Processing School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta GA USA

Motivated by the success of support vector regression (SVR) in blind image deconvolution, we apply SVR to single-frame super-resolution. Initial results show that even when trained on as little as a single image, SVR ... 详细信息

关键词： image resolution signal resolution Deconvolution Interpolation Low pass filters Discrete wavelet transforms PSNR High-resolution imaging Filtering Discrete cosine transforms

Radiomics analysis of baseline F-FDG PET/CT images for improved prognosis in nasopharyngeal carcinoma 15

学校读者我要写书评

暂无评论

Radiomics analysis of baseline F-FDG PET/CT images for impro...

15th IEEE International Symposium on Biomedical Imaging, ISBI 2018

作者： Lv, Wenbing Yuan, Qingyu Wang, Quanshi Ma, Jianhua Feng, Qianjin Chen, Wufan Rahmim, Arman Lu, Lijun School of Biomedical Engineering and Guangdong Provincial Key Laboratory of Medical Image Processing Southern Medical University China Nanfang PET Center Nanfang Hospital Southern Medical University China Department of Radiology Johns Hopkins University United States Department of Electrical and Computer Engineering Johns Hopkins University United States

ISBN: (纸本)9781538636367

The purpose of this study is to investigate the prognostic performance of radiomics features on nasopharyngeal carcinoma (NPC) patients imaged with baseline 18F-FDG PET/CT. 128 NPC patients were retrospectively enrolled with 3348 radiomics features and 13 clinical features. Kaplan-Meier analysis was used to estimate progression-free survival (PFS), and log-rank test was used to screen the significant features. Cox proportional hazards regression modal with forward stepwise feature selection was adopted to identify independent predictors of PFS. 24 radiomics features and 8 clinical features were found to be significantly associated with PFS in univariate analysis. Radiomics features HGZE-GLSZM-HLL-32 (p=0.0061, HR: 0.66, 95%CI: 0.49-0.89), as well as clinical features N-stage, M-stage, antibody VCA-IgA and platelet count (PLT) retained the independent prognostic significance for PFS in multivariate analysis. Overall, radiomics features can provide complementary prognostic information for NPC patients imaged with baseline 18F-FDG PET/CT compared to clinical features. © 2018 IEEE.

关键词： Multivariant analysis