38 results on '"Kim, Hyunwoo J."'
Search Results
2. Randomly shuffled convolution for self-supervised representation learning
- Author
-
Oh, Youngjin, Jeon, Minkyu, Ko, Dohwan, and Kim, Hyunwoo J.
- Published
- 2023
- Full Text
- View/download PDF
3. Graph Transformer Networks: Learning meta-path graphs to improve GNNs
- Author
-
Yun, Seongjun, Jeong, Minbyul, Yoo, Sungdong, Lee, Seunghun, Yi, Sean S., Kim, Raehyun, Kang, Jaewoo, and Kim, Hyunwoo J.
- Published
- 2022
- Full Text
- View/download PDF
4. Single cell lineage reconstruction using distance-based algorithms and the R package, DCLEAR
- Author
-
Gong, Wuming, Kim, Hyunwoo J., Garry, Daniel J., and Kwak, Il-Youp
- Published
- 2022
- Full Text
- View/download PDF
5. Self‐Powering Sensory Device with Multi‐Spectrum Image Realization for Smart Indoor Environments.
- Author
-
Kim, Tae Hyuk, Yu, Byoung‐Soo, Ko, Hyun Woo, Park, Na Won, Saeed, Muhammad Ahsan, Ahn, Jongtae, Jo, Suyeon, Kim, Dae‐Yeon, Yoon, Seon Kyu, Lee, Kwang‐Hoon, Jeong, Sang Young, Woo, Han Young, Kim, Hyunwoo J., Kim, Tae Geun, Park, JaeHong, Park, Min‐Chul, Hwang, Do Kyung, and Shim, Jae Won
- Published
- 2024
- Full Text
- View/download PDF
6. MELTR: Meta Loss Transformer for Learning to Fine-tune Video Foundation Models
- Author
-
Ko, Dohwan, Choi, Joonmyung, Choi, Hyeong Kyu, On, Kyoung-Woon, Roh, Byungseok, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition - Abstract
Foundation models have shown outstanding performance and generalization capabilities across domains. Since most studies on foundation models mainly focus on the pretraining phase, a naive strategy to minimize a single task-specific loss is adopted for fine-tuning. However, such fine-tuning methods do not fully leverage other losses that are potentially beneficial for the target task. Therefore, we propose MEta Loss TRansformer (MELTR), a plug-in module that automatically and non-linearly combines various loss functions to aid learning the target task via auxiliary learning. We formulate the auxiliary learning as a bi-level optimization problem and present an efficient optimization algorithm based on Approximate Implicit Differentiation (AID). For evaluation, we apply our framework to various video foundation models (UniVL, Violet and All-in-one), and show significant performance gain on all four downstream tasks: text-to-video retrieval, video question answering, video captioning, and multi-modal sentiment analysis. Our qualitative analyses demonstrate that MELTR adequately `transforms' individual loss functions and `melts' them into an effective unified loss. Code is available at https://github.com/mlvlab/MELTR., Accepted paper at CVPR 2023
- Published
- 2023
7. k-SALSA: k-anonymous synthetic averaging of retinal images via local style alignment
- Author
-
Jeon, Minkyu, Park, Hyeonjin, Kim, Hyunwoo J., Morley, Michael, and Cho, Hyunghoon
- Subjects
FOS: Computer and information sciences ,Computer Science - Cryptography and Security ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,Cryptography and Security (cs.CR) - Abstract
The application of modern machine learning to retinal image analyses offers valuable insights into a broad range of human health conditions beyond ophthalmic diseases. Additionally, data sharing is key to fully realizing the potential of machine learning models by providing a rich and diverse collection of training data. However, the personally-identifying nature of retinal images, encompassing the unique vascular structure of each individual, often prevents this data from being shared openly. While prior works have explored image de-identification strategies based on synthetic averaging of images in other domains (e.g. facial images), existing techniques face difficulty in preserving both privacy and clinical utility in retinal images, as we demonstrate in our work. We therefore introduce k-SALSA, a generative adversarial network (GAN)-based framework for synthesizing retinal fundus images that summarize a given private dataset while satisfying the privacy notion of k-anonymity. k-SALSA brings together state-of-the-art techniques for training and inverting GANs to achieve practical performance on retinal images. Furthermore, k-SALSA leverages a new technique, called local style alignment, to generate a synthetic average that maximizes the retention of fine-grain visual patterns in the source images, thus improving the clinical utility of the generated images. On two benchmark datasets of diabetic retinopathy (EyePACS and APTOS), we demonstrate our improvement upon existing methods with respect to image fidelity, classification performance, and mitigation of membership inference attacks. Our work represents a step toward broader sharing of retinal images for scientific collaboration. Code is available at https://github.com/hcholab/k-salsa., European Conference on Computer Vision (ECCV), 2022
- Published
- 2023
8. Semantic-aware Occlusion Filtering Neural Radiance Fields in the Wild
- Author
-
Lee, Jaewon, Kim, Injae, Heo, Hwan, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Artificial Intelligence (cs.AI) ,Computer Science - Graphics ,Computer Science - Artificial Intelligence ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,Graphics (cs.GR) ,Machine Learning (cs.LG) - Abstract
We present a learning framework for reconstructing neural scene representations from a small number of unconstrained tourist photos. Since each image contains transient occluders, decomposing the static and transient components is necessary to construct radiance fields with such in-the-wild photographs where existing methods require a lot of training data. We introduce SF-NeRF, aiming to disentangle those two components with only a few images given, which exploits semantic information without any supervision. The proposed method contains an occlusion filtering module that predicts the transient color and its opacity for each pixel, which enables the NeRF model to solely learn the static scene representation. This filtering module learns the transient phenomena guided by pixel-wise semantic features obtained by a trainable image encoder that can be trained across multiple scenes to learn the prior of transient objects. Furthermore, we present two techniques to prevent ambiguous decomposition and noisy results of the filtering module. We demonstrate that our method outperforms state-of-the-art novel view synthesis methods on Phototourism dataset in a few-shot setting., 11 pages, 5 figures
- Published
- 2023
9. Robust Camera Pose Refinement for Multi-Resolution Hash Encoding
- Author
-
Heo, Hwan, Kim, Taekyung, Lee, Jiyoung, Lee, Jaewon, Kim, Soohyun, Kim, Hyunwoo J., and Kim, Jin-Hwa
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Artificial Intelligence (cs.AI) ,Computer Science - Graphics ,Computer Science - Artificial Intelligence ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,Graphics (cs.GR) ,Machine Learning (cs.LG) - Abstract
Multi-resolution hash encoding has recently been proposed to reduce the computational cost of neural renderings, such as NeRF. This method requires accurate camera poses for the neural renderings of given scenes. However, contrary to previous methods jointly optimizing camera poses and 3D scenes, the naive gradient-based camera pose refinement method using multi-resolution hash encoding severely deteriorates performance. We propose a joint optimization algorithm to calibrate the camera pose and learn a geometric representation using efficient multi-resolution hash encoding. Showing that the oscillating gradient flows of hash encoding interfere with the registration of camera poses, our method addresses the issue by utilizing smooth interpolation weighting to stabilize the gradient oscillation for the ray samplings across hash grids. Moreover, the curriculum training procedure helps to learn the level-wise hash encoding, further increasing the pose refinement. Experiments on the novel-view synthesis datasets validate that our learning frameworks achieve state-of-the-art performance and rapid convergence of neural rendering, even when initial camera poses are unknown.
- Published
- 2023
10. Domain Generalization Emerges from Dreaming
- Author
-
Heo, Hwan, Oh, Youngjin, Lee, Jaewon, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Artificial Intelligence (cs.AI) ,Computer Science - Artificial Intelligence ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,Machine Learning (cs.LG) - Abstract
Recent studies have proven that DNNs, unlike human vision, tend to exploit texture information rather than shape. Such texture bias is one of the factors for the poor generalization performance of DNNs. We observe that the texture bias negatively affects not only in-domain generalization but also out-of-distribution generalization, i.e., Domain Generalization. Motivated by the observation, we propose a new framework to reduce the texture bias of a model by a novel optimization-based data augmentation, dubbed Stylized Dream. Our framework utilizes adaptive instance normalization (AdaIN) to augment the style of an original image yet preserve the content. We then adopt a regularization loss to predict consistent outputs between Stylized Dream and original images, which encourages the model to learn shape-based representations. Extensive experiments show that the proposed method achieves state-of-the-art performance in out-of-distribution settings on public benchmark datasets: PACS, VLCS, OfficeHome, TerraIncognita, and DomainNet., 23 pages, 4 figures
- Published
- 2023
11. SageMix: Saliency-Guided Mixup for Point Clouds
- Author
-
Lee, Sanghyeok, Jeon, Minkyu, Kim, Injae, Xiong, Yunyang, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,Machine Learning (cs.LG) - Abstract
Data augmentation is key to improving the generalization ability of deep learning models. Mixup is a simple and widely-used data augmentation technique that has proven effective in alleviating the problems of overfitting and data scarcity. Also, recent studies of saliency-aware Mixup in the image domain show that preserving discriminative parts is beneficial to improving the generalization performance. However, these Mixup-based data augmentations are underexplored in 3D vision, especially in point clouds. In this paper, we propose SageMix, a saliency-guided Mixup for point clouds to preserve salient local structures. Specifically, we extract salient regions from two point clouds and smoothly combine them into one continuous shape. With a simple sequential sampling by re-weighted saliency scores, SageMix preserves the local structure of salient regions. Extensive experiments demonstrate that the proposed method consistently outperforms existing Mixup methods in various benchmark point cloud datasets. With PointNet++, our method achieves an accuracy gain of 2.6% and 4.0% over standard training in 3D Warehouse dataset (MN40) and ScanObjectNN, respectively. In addition to generalization performance, SageMix improves robustness and uncertainty calibration. Moreover, when adopting our method to various tasks including part segmentation and standard 2D image classification, our method achieves competitive performance., Neural Information Processing Systems (NeurIPS), 2022
- Published
- 2022
12. Deformable Graph Transformer
- Author
-
Park, Jinyoung, Yun, Seongjun, Park, Hyeonjin, Kang, Jaewoo, Jeong, Jisu, Kim, Kyung-Min, Ha, Jung-woo, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Machine Learning (cs.LG) - Abstract
Transformer-based models have recently shown success in representation learning on graph-structured data beyond natural language processing and computer vision. However, the success is limited to small-scale graphs due to the drawbacks of full dot-product attention on graphs such as the quadratic complexity with respect to the number of nodes and message aggregation from enormous irrelevant nodes. To address these issues, we propose Deformable Graph Transformer (DGT) that performs sparse attention via dynamically sampled relevant nodes for efficiently handling large-scale graphs with a linear complexity in the number of nodes. Specifically, our framework first constructs multiple node sequences with various criteria to consider both structural and semantic proximity. Then, combining with our learnable Katz Positional Encodings, the sparse attention is applied to the node sequences for learning node representations with a significantly reduced computational cost. Extensive experiments demonstrate that our DGT achieves state-of-the-art performance on 7 graph benchmark datasets with 2.5 - 449 times less computational cost compared to transformer-based graph models with full attention., 16 pages, 3 figures
- Published
- 2022
13. Neo-GNNs: Neighborhood Overlap-aware Graph Neural Networks for Link Prediction
- Author
-
Yun, Seongjun, Kim, Seoyoon, Lee, Junhyun, Kang, Jaewoo, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Artificial Intelligence (cs.AI) ,Computer Science - Artificial Intelligence ,Machine Learning (cs.LG) - Abstract
Graph Neural Networks (GNNs) have been widely applied to various fields for learning over graph-structured data. They have shown significant improvements over traditional heuristic methods in various tasks such as node classification and graph classification. However, since GNNs heavily rely on smoothed node features rather than graph structure, they often show poor performance than simple heuristic methods in link prediction where the structural information, e.g., overlapped neighborhoods, degrees, and shortest paths, is crucial. To address this limitation, we propose Neighborhood Overlap-aware Graph Neural Networks (Neo-GNNs) that learn useful structural features from an adjacency matrix and estimate overlapped neighborhoods for link prediction. Our Neo-GNNs generalize neighborhood overlap-based heuristic methods and handle overlapped multi-hop neighborhoods. Our extensive experiments on Open Graph Benchmark datasets (OGB) demonstrate that Neo-GNNs consistently achieve state-of-the-art performance in link prediction. Our code is publicly available at https://github.com/seongjunyun/Neo_GNNs.
- Published
- 2022
14. Consistency Learning via Decoding Path Augmentation for Transformers in Human Object Interaction Detection
- Author
-
Park, Jihwan, Lee, SeungJun, Heo, Hwan, Choi, Hyeong Kyu, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition - Abstract
Human-Object Interaction detection is a holistic visual recognition task that entails object detection as well as interaction classification. Previous works of HOI detection has been addressed by the various compositions of subset predictions, e.g., Image -> HO -> I, Image -> HI -> O. Recently, transformer based architecture for HOI has emerged, which directly predicts the HOI triplets in an end-to-end fashion (Image -> HOI). Motivated by various inference paths for HOI detection, we propose cross-path consistency learning (CPC), which is a novel end-to-end learning strategy to improve HOI detection for transformers by leveraging augmented decoding paths. CPC learning enforces all the possible predictions from permuted inference sequences to be consistent. This simple scheme makes the model learn consistent representations, thereby improving generalization without increasing model capacity. Our experiments demonstrate the effectiveness of our method, and we achieved significant improvement on V-COCO and HICO-DET compared to the baseline models. Our code is available at https://github.com/mlvlab/CPChoi., CVPR2022 accepted
- Published
- 2022
15. Metropolis-Hastings Data Augmentation for Graph Neural Networks
- Author
-
Park, Hyeonjin, Lee, Seunghun, Kim, Sihyeon, Park, Jinyoung, Jeong, Jisu, Kim, Kyung-Min, Ha, Jung-Woo, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Machine Learning (cs.LG) - Abstract
Graph Neural Networks (GNNs) often suffer from weak-generalization due to sparsely labeled data despite their promising results on various graph-based tasks. Data augmentation is a prevalent remedy to improve the generalization ability of models in many domains. However, due to the non-Euclidean nature of data space and the dependencies between samples, designing effective augmentation on graphs is challenging. In this paper, we propose a novel framework Metropolis-Hastings Data Augmentation (MH-Aug) that draws augmented graphs from an explicit target distribution for semi-supervised learning. MH-Aug produces a sequence of augmented graphs from the target distribution enables flexible control of the strength and diversity of augmentation. Since the direct sampling from the complex target distribution is challenging, we adopt the Metropolis-Hastings algorithm to obtain the augmented samples. We also propose a simple and effective semi-supervised learning strategy with generated samples from MH-Aug. Our extensive experiments demonstrate that MH-Aug can generate a sequence of samples according to the target distribution to significantly improve the performance of GNNs., 10 pages, 5 figures
- Published
- 2022
16. Self-supervised Auxiliary Learning with Meta-paths for Heterogeneous Graphs
- Author
-
Hwang, Dasol, Park, Jinyoung, Kwon, Sunyoung, Kim, Kyung-Min, Ha, Jung-Woo, and Kim, Hyunwoo J.
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Statistics - Machine Learning ,Machine Learning (stat.ML) ,Machine Learning (cs.LG) - Abstract
Graph neural networks have shown superior performance in a wide range of applications providing a powerful representation of graph-structured data. Recent works show that the representation can be further improved by auxiliary tasks. However, the auxiliary tasks for heterogeneous graphs, which contain rich semantic information with various types of nodes and edges, have less explored in the literature. In this paper, to learn graph neural networks on heterogeneous graphs we propose a novel self-supervised auxiliary learning method using meta-paths, which are composite relations of multiple edge types. Our proposed method is learning to learn a primary task by predicting meta-paths as auxiliary tasks. This can be viewed as a type of meta-learning. The proposed method can identify an effective combination of auxiliary tasks and automatically balance them to improve the primary task. Our methods can be applied to any graph neural networks in a plug-in manner without manual labeling or additional data. The experiments demonstrate that the proposed method consistently improves the performance of link prediction and node classification on heterogeneous graphs., Neural Information Processing Systems (NeurIPS), 2020
- Published
- 2020
17. Graph Transformer Networks
- Author
-
Yun, Seongjun, Jeong, Minbyul, Kim, Raehyun, Kang, Jaewoo, and Kim, Hyunwoo J.
- Subjects
Social and Information Networks (cs.SI) ,FOS: Computer and information sciences ,Computer Science - Machine Learning ,Statistics - Machine Learning ,Machine Learning (stat.ML) ,Computer Science - Social and Information Networks ,Machine Learning (cs.LG) ,MathematicsofComputing_DISCRETEMATHEMATICS - Abstract
Graph neural networks (GNNs) have been widely used in representation learning on graphs and achieved state-of-the-art performance in tasks such as node classification and link prediction. However, most existing GNNs are designed to learn node representations on the fixed and homogeneous graphs. The limitations especially become problematic when learning representations on a misspecified graph or a heterogeneous graph that consists of various types of nodes and edges. In this paper, we propose Graph Transformer Networks (GTNs) that are capable of generating new graph structures, which involve identifying useful connections between unconnected nodes on the original graph, while learning effective node representation on the new graphs in an end-to-end fashion. Graph Transformer layer, a core layer of GTNs, learns a soft selection of edge types and composite relations for generating useful multi-hop connections so-called meta-paths. Our experiments show that GTNs learn new graph structures, based on data and tasks without domain knowledge, and yield powerful node representation via convolution on the new graphs. Without domain-specific graph preprocessing, GTNs achieved the best performance in all three benchmark node classification tasks against the state-of-the-art methods that require pre-defined meta-paths from domain knowledge., Neural Information Processing Systems (NeurIPS), 2019
- Published
- 2019
18. ANTNets: Mobile Convolutional Neural Networks for Resource Efficient Image Classification
- Author
-
Xiong, Yunyang, Kim, Hyunwoo J., and Hedau, Varsha
- Subjects
FOS: Computer and information sciences ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition - Abstract
Deep convolutional neural networks have achieved remarkable success in computer vision. However, deep neural networks require large computing resources to achieve high performance. Although depthwise separable convolution can be an efficient module to approximate a standard convolution, it often leads to reduced representational power of networks. In this paper, under budget constraints such as computational cost (MAdds) and the parameter count, we propose a novel basic architectural block, ANTBlock. It boosts the representational power by modeling, in a high dimensional space, interdependency of channels between a depthwise convolution layer and a projection layer in the ANTBlocks. Our experiments show that ANTNet built by a sequence of ANTBlocks, consistently outperforms state-of-the-art low-cost mobile convolutional neural networks across multiple datasets. On CIFAR100, our model achieves 75.7% top-1 accuracy, which is 1.5% higher than MobileNetV2 with 8.3% fewer parameters and 19.6% less computational cost. On ImageNet, our model achieves 72.8% top-1 accuracy, which is 0.8% improvement, with 157.7ms (20% faster) on iPhone 5s over MobileNetV2., CVPR 2019 Workshop on Efficient Deep Learning for Computer Vision (Oral)
- Published
- 2019
19. Manifold-valued Dirichlet Processes
- Author
-
Kim, Hyunwoo J., Xu, Jia, Vemuri, Baba C., and Singh, Vikas
- Subjects
Mathematics::Differential Geometry ,Article - Abstract
Statistical models for manifold-valued data permit capturing the intrinsic nature of the curved spaces in which the data lie and have been a topic of research for several decades. Typically, these formulations use geodesic curves and distances defined locally for most cases — this makes it hard to design parametric models globally on smooth manifolds. Thus, most (manifold specific) parametric models available today assume that the data lie in a small neighborhood on the manifold. To address this ‘locality’ problem, we propose a novel nonparametric model which unifies multivariate general linear models (MGLMs) using multiple tangent spaces. Our framework generalizes existing work on (both Euclidean and non-Euclidean) general linear models providing a recipe to globally extend the locally-defined parametric models (using a mixture of local models). By grouping observations into sub-populations at multiple tangent spaces, our method provides insights into the hidden structure (geodesic relationships) in the data. This yields a framework to group observations and discover geodesic relationships between covariates X and manifold-valued responses Y, which we call Dirichlet process mixtures of multivariate general linear models (DP-MGLM) on Riemannian manifolds. Finally, we present proof of concept experiments to validate our model.
- Published
- 2015
20. LOCALIZING DIFFERENTIALLY EVOLVING COVARIANCE STRUCTURES VIA SCAN STATISTICS.
- Author
-
MEHTA, RONAK, KIM, HYUNWOO J., SHULEI WANG, JOHNSON, STERLING C., MING YUAN, and SINGH, VIKAS
- Subjects
GRAPHICAL modeling (Statistics) ,STATISTICS - Abstract
Recent results in coupled or temporal graphical models offer schemes for estimating the relationship structure between features when the data come from related (but distinct) longitudinal sources. A novel application of these ideas is for analyzing group-level differences, i.e., in identifying if trends of estimated objects (e.g., covariance or precision matrices) are different across disparate conditions (e.g., gender or disease). Often, poor effect sizes make detecting the differential signal over the full set of features difficult: for example, dependencies between only a subset of features may manifest differently across groups. In this work, we first give a parametric model for estimating trends in the space of SPD matrices as a function of one or more covariates. We then generalize scan statistics to graph structures, to search over distinct subsets of features (graph partitions) whose temporal dependency structure may show statistically significant groupwise differences. We theoretically analyze the Family Wise Error Rate (FWER) and bounds on Type 1 and Type 2 error. Evaluating on U.S. census data, we identify groups of states with cultural and legal overlap related to baby name trends and drug usage. On a cohort of individuals with risk factors for Alzheimer's disease (but otherwise cognitively healthy), we find scientifically interesting group differences where the default analysis, i.e., models estimated on the full graph, do not survive reasonable significance thresholds. [ABSTRACT FROM AUTHOR]
- Published
- 2019
- Full Text
- View/download PDF
21. COVER SONG IDENTIFICATION WITH METRIC LEARNING USING DISTANCE AS A FEATURE.
- Author
-
Hoon Heo, Kim, Hyunwoo J., Wan Soo Kim, and Kyogu Lee
- Subjects
COVER versions ,MUSICAL meter & rhythm ,POPULAR music genres ,FOURIER transforms ,PRINCIPAL components analysis - Abstract
Most of cover song identification algorithms are based on the pairwise (dis)similarity between two songs which are represented by harmonic features such as chroma, and therefore the choice of a distance measure and a feature has a significant impact on performance. Furthermore, since the similarity measure is query-dependent, it cannot represent an absolute distance measure. In this paper, we present a novel approach to tackle the cover song identification problem from a new perspective. We first construct a set of core songs, and represent each song in a high-dimensional space where each dimension indicates the pairwise distance between the given song and the other in the pre-defined core set. There are several advantages to this. First, using a number of reference songs in the core set, we make the most of relative distances to many other songs. Second, as all songs are transformed into the same high-dimensional space, kernel methods and metric learning are exploited for distance computation. Third, our approach does not depend on the computation method for the pairwise distance, and thus can use any existing algorithms. Experimental results confirm that the proposed approach achieved a large performance gain compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]
- Published
- 2017
22. Latent Variable Graphical Model Selection Using Harmonic Analysis: Applications to the Human Connectome Project (HCP).
- Author
-
Kim, Won Hwa, Kim, Hyunwoo J., Adluru, Nagesh, and Singh, Vikas
- Published
- 2016
- Full Text
- View/download PDF
23. Abundant Inverse Regression Using Sufficient Reduction and Its Applications.
- Author
-
Kim, Hyunwoo J., Smith, Brandon M., Adluru, Nagesh, Dyer, Charles R., Johnson, Sterling C., and Singh, Vikas
- Published
- 2016
- Full Text
- View/download PDF
24. Canonical Correlation Analysis on SPD(n) Manifolds.
- Author
-
Kim, Hyunwoo J., Adluru, Nagesh, Bendlin, Barbara B., Johnson, Sterling C., Vemuri, Baba C., and Singh, Vikas
- Published
- 2016
- Full Text
- View/download PDF
25. Interpolation on the Manifold of K Component GMMs.
- Author
-
Kim, Hyunwoo J., Adluru, Nagesh, Banerjee, Monami, Vemuri, Baba C., and Singh, Vikas
- Published
- 2015
- Full Text
- View/download PDF
26. Predicting Unseen Labels Using Label Hierarchies in Large-Scale Multi-label Learning.
- Author
-
Nam, Jinseok, Loza Mencía, Eneldo, Kim, Hyunwoo J., and Fürnkranz, Johannes
- Published
- 2015
- Full Text
- View/download PDF
27. Multivariate General Linear Models (MGLM) on Riemannian Manifolds with Applications to Statistical Analysis of Diffusion Weighted Images.
- Author
-
Kim, Hyunwoo J., Bendlin, Barbara B., Adluru, Nagesh, Collins, Maxwell D., Chung, Moo K., Johnson, Sterling C., Davidson, Richard J., and Singh, Vikas
- Published
- 2014
- Full Text
- View/download PDF
28. Canonical Correlation Analysis on Riemannian Manifolds and Its Applications.
- Author
-
Kim, Hyunwoo J., Adluru, Nagesh, Bendlin, Barbara B., Johnson, Sterling C., Vemuri, Baba C., and Singh, Vikas
- Published
- 2014
- Full Text
- View/download PDF
29. k -SALSA: k -anonymous synthetic averaging of retinal images via local style alignment.
- Author
-
Jeon M, Park H, Kim HJ, Morley M, and Cho H
- Abstract
The application of modern machine learning to retinal image analyses offers valuable insights into a broad range of human health conditions beyond ophthalmic diseases. Additionally, data sharing is key to fully realizing the potential of machine learning models by providing a rich and diverse collection of training data. However, the personallyidentifying nature of retinal images, encompassing the unique vascular structure of each individual, often prevents this data from being shared openly. While prior works have explored image de-identification strategies based on synthetic averaging of images in other domains (e.g. facial images), existing techniques face difficulty in preserving both privacy and clinical utility in retinal images, as we demonstrate in our work. We therefore introduce k -SALSA, a generative adversarial network (GAN)-based framework for synthesizing retinal fundus images that summarize a given private dataset while satisfying the privacy notion of k -anonymity. k -SALSA brings together state-of-the-art techniques for training and inverting GANs to achieve practical performance on retinal images. Furthermore, k -SALSA leverages a new technique, called local style alignment, to generate a synthetic average that maximizes the retention of fine-grain visual patterns in the source images, thus improving the clinical utility of the generated images. On two benchmark datasets of diabetic retinopathy (EyePACS and APTOS), we demonstrate our improvement upon existing methods with respect to image fidelity, classification performance, and mitigation of membership inference attacks. Our work represents a step toward broader sharing of retinal images for scientific collaboration. Code is available at https://github.com/hcholab/k-salsa.
- Published
- 2022
- Full Text
- View/download PDF
30. Sampling-free Uncertainty Estimation in Gated Recurrent Units with Applications to Normative Modeling in Neuroimaging.
- Author
-
Hwang SJ, Mehta RR, Kim HJ, Johnson SC, and Singh V
- Abstract
There has recently been a concerted effort to derive mechanisms in vision and machine learning systems to offer uncertainty estimates of the predictions they make. Clearly, there are benefits to a system that is not only accurate but also has a sense for when it is not. Existing proposals center around Bayesian interpretations of modern deep architectures - these are effective but can often be computationally demanding. We show how classical ideas in the literature on exponential families on probabilistic networks provide an excellent starting point to derive uncertainty estimates in Gated Recurrent Units (GRU). Our proposal directly quantifies uncertainty deterministically , without the need for costly sampling-based estimation. We show that while uncertainty is quite useful by itself in computer vision and machine learning, we also demonstrate that it can play a key role in enabling statistical analysis with deep networks in neuroimaging studies with normative modeling methods. To our knowledge, this is the first result describing sampling-free uncertainty estimation for powerful sequential models such as GRUs.
- Published
- 2019
31. On Training Deep 3D CNN Models with Dependent Samples in Neuroimaging.
- Author
-
Xiong Y, Kim HJ, Tangirala B, Mehta R, Johnson SC, and Singh V
- Abstract
There is much interest in developing algorithms based on 3D convolutional neural networks (CNNs) for performing regression and classification with brain imaging data and more generally, with biomedical imaging data. A standard assumption in learning is that the training samples are independently drawn from the underlying distribution. In computer vision, where we have millions of training examples, this assumption is violated but the empirical performance may remain satisfactory. But in many biomedical studies with just a few hundred training examples, one often has multiple samples per participant and/or data may be curated by pooling datasets from a few different institutions. Here, the violation of the independent samples assumption turns out to be more significant, especially in small-to-medium sized datasets. Motivated by this need, we show how 3D CNNs can be modified to deal with dependent samples. We show that even with standard 3D CNNs, there is value in augmenting the network to exploit information regarding dependent samples. We present empirical results for predicting cognitive trajectories (slope and intercept) from morphometric change images derived from multiple time points. With terms which encode dependency between samples in the model, we get consistent improvements over a strong baseline which ignores such knowledge.
- Published
- 2019
- Full Text
- View/download PDF
32. Efficient Relative Attribute Learning using Graph Neural Networks.
- Author
-
Meng Z, Adluru N, Kim HJ, Fung G, and Singh V
- Abstract
A sizable body of work on relative attributes provides evidence that relating pairs of images along a continuum of strength pertaining to a visual attribute yields improvements in a variety of vision tasks. In this paper, we show how emerging ideas in graph neural networks can yield a solution to various problems that broadly fall under relative attribute learning. Our main idea is the observation that relative attribute learning naturally benefits from exploiting the graph of dependencies among the different relative attributes of images, especially when only partial ordering is provided at training time. We use message passing to perform end to end learning of the image representations, their relationships as well as the interplay between different attributes. Our experiments show that this simple framework is effective in achieving competitive accuracy with specialized methods for both relative attribute learning and binary attribute prediction, while relaxing the requirements on the training data and/or the number of parameters, or both.
- Published
- 2018
- Full Text
- View/download PDF
33. Riemannian Variance Filtering: An Independent Filtering Scheme for Statistical Tests on Manifold-valued Data.
- Author
-
Zheng L, Kim HJ, Adluru N, Newton MA, and Singh V
- Abstract
Performing large scale hypothesis testing on brain imaging data to identify group-wise differences (e.g., between healthy and diseased subjects) typically leads to a large number of tests (one per voxel). Multiple testing adjustment (or correction) is necessary to control false positives, which may lead to lower detection power in detecting true positives. Motivated by the use of so-called "independent filtering" techniques in statistics (for genomics applications), this paper investigates the use of independent filtering for manifold-valued data (e.g., Diffusion Tensor Imaging, Cauchy Deformation Tensors) which are broadly used in neuroimaging studies. Inspired by the concept of variance of a Riemannian Gaussian distribution, a type of non-specific data-dependent Riemannian variance filter is proposed. In practice, the filter will select a subset of the full set of voxels for performing the statistical test, leading to a more appropriate multiple testing correction. Our experiments on synthetic/simulated manifold-valued data show that the detection power is improved when the statistical tests are performed on the voxel locations that "pass" the filter. Given the broadening scope of applications where manifold-valued data are utilized, the scheme can serve as a general feature selection scheme.
- Published
- 2017
- Full Text
- View/download PDF
34. Riemannian Nonlinear Mixed Effects Models: Analyzing Longitudinal Deformations in Neuroimaging.
- Author
-
Kim HJ, Adluru N, Suri H, Vemuri BC, Johnson SC, and Singh V
- Abstract
Statistical machine learning models that operate on manifold-valued data are being extensively studied in vision, motivated by applications in activity recognition, feature tracking and medical imaging. While non-parametric methods have been relatively well studied in the literature, efficient formulations for parametric models (which may offer benefits in small sample size regimes) have only emerged recently. So far, manifold-valued regression models (such as geodesic regression) are restricted to the analysis of cross-sectional data, i.e., the so-called "fixed effects" in statistics. But in most "longitudinal analysis" (e.g., when a participant provides multiple measurements, over time) the application of fixed effects models is problematic. In an effort to answer this need, this paper generalizes non-linear mixed effects model to the regime where the response variable is manifold-valued, i.e., f : R
d → ℳ. We derive the underlying model and estimation schemes and demonstrate the immediate benefits such a model can provide - both for group level and individual level analysis - on longitudinal brain imaging data. The direct consequence of our results is that longitudinal analysis of manifold-valued measurements (especially, the symmetric positive definite manifold) can be conducted in a computationally tractable manner.- Published
- 2017
- Full Text
- View/download PDF
35. Latent Variable Graphical Model Selection using Harmonic Analysis: Applications to the Human Connectome Project (HCP).
- Author
-
Kim WH, Kim HJ, Adluru N, and Singh V
- Abstract
A major goal of imaging studies such as the (ongoing) Human Connectome Project (HCP) is to characterize the structural network map of the human brain and identify its associations with covariates such as genotype, risk factors, and so on that correspond to an individual. But the set of image derived measures and the set of covariates are both large, so we must first estimate a 'parsimonious' set of relations between the measurements. For instance, a Gaussian graphical model will show conditional independences between the random variables, which can then be used to setup specific downstream analyses. But most such data involve a large list of 'latent' variables that remain unobserved, yet affect the 'observed' variables sustantially. Accounting for such latent variables is not directly addressed by standard precision matrix estimation, and is tackled via highly specialized optimization methods. This paper offers a unique harmonic analysis view of this problem. By casting the estimation of the precision matrix in terms of a composition of low-frequency latent variables and high-frequency sparse terms, we show how the problem can be formulated using a new wavelet-type expansion in non-Euclidean spaces. Our formulation poses the estimation problem in the frequency space and shows how it can be solved by a simple sub-gradient scheme. We provide a set of scientific results on ~500 scans from the recently released HCP data where our algorithm recovers highly interpretable and sparse conditional dependencies between brain connectivity pathways and well-known covariates.
- Published
- 2016
- Full Text
- View/download PDF
36. Interpolation on the manifold of K component GMMs.
- Author
-
Kim HJ, Adluru N, Banerjee M, Vemuri BC, and Singh V
- Abstract
Probability density functions (PDFs) are fundamental objects in mathematics with numerous applications in computer vision, machine learning and medical imaging. The feasibility of basic operations such as computing the distance between two PDFs and estimating a mean of a set of PDFs is a direct function of the representation we choose to work with. In this paper, we study the Gaussian mixture model (GMM) representation of the PDFs motivated by its numerous attractive features. (1) GMMs are arguably more interpretable than, say, square root parameterizations (2) the model complexity can be explicitly controlled by the number of components and (3) they are already widely used in many applications. The main contributions of this paper are numerical algorithms to enable basic operations on such objects that strictly respect their underlying geometry. For instance, when operating with a set of K component GMMs, a first order expectation is that the result of simple operations like interpolation and averaging should provide an object that is also a K component GMM. The literature provides very little guidance on enforcing such requirements systematically. It turns out that these tasks are important internal modules for analysis and processing of a field of ensemble average propagators (EAPs), common in diffusion weighted magnetic resonance imaging. We provide proof of principle experiments showing how the proposed algorithms for interpolation can facilitate statistical analysis of such data, essential to many neuroimaging studies. Separately, we also derive interesting connections of our algorithm with functional spaces of Gaussians, that may be of independent interest.
- Published
- 2015
- Full Text
- View/download PDF
37. Manifold-valued Dirichlet Processes.
- Author
-
Kim HJ, Xu J, Vemuri BC, and Singh V
- Abstract
Statistical models for manifold-valued data permit capturing the intrinsic nature of the curved spaces in which the data lie and have been a topic of research for several decades. Typically, these formulations use geodesic curves and distances defined locally for most cases - this makes it hard to design parametric models globally on smooth manifolds. Thus, most (manifold specific) parametric models available today assume that the data lie in a small neighborhood on the manifold. To address this 'locality' problem, we propose a novel nonparametric model which unifies multivariate general linear models (MGLMs) using multiple tangent spaces. Our framework generalizes existing work on (both Euclidean and non-Euclidean) general linear models providing a recipe to globally extend the locally-defined parametric models (using a mixture of local models). By grouping observations into sub-populations at multiple tangent spaces, our method provides insights into the hidden structure (geodesic relationships) in the data. This yields a framework to group observations and discover geodesic relationships between covariates X and manifold-valued responses Y , which we call Dirichlet process mixtures of multivariate general linear models (DP-MGLM) on Riemannian manifolds. Finally, we present proof of concept experiments to validate our model.
- Published
- 2015
38. Multivariate General Linear Models (MGLM) on Riemannian Manifolds with Applications to Statistical Analysis of Diffusion Weighted Images.
- Author
-
Kim HJ, Adluru N, Collins MD, Chung MK, Bendlin BB, Johnson SC, Davidson RJ, and Singh V
- Abstract
Linear regression is a parametric model which is ubiquitous in scientific analysis. The classical setup where the observations and responses, i.e., ( x
i , yi ) pairs, are Euclidean is well studied. The setting where yi is manifold valued is a topic of much interest, motivated by applications in shape analysis, topic modeling, and medical imaging. Recent work gives strategies for max-margin classifiers, principal components analysis, and dictionary learning on certain types of manifolds. For parametric regression specifically, results within the last year provide mechanisms to regress one real-valued parameter, xi ∈ R , against a manifold-valued variable, yi ∈ . We seek to substantially extend the operating range of such methods by deriving schemes for multivariate multiple linear regression -a manifold-valued dependent variable against multiple independent variables, i.e., f : Rn → . Our variational algorithm efficiently solves for multiple geodesic bases on the manifold concurrently via gradient updates. This allows us to answer questions such as: what is the relationship of the measurement at voxel y to disease when conditioned on age and gender. We show applications to statistical analysis of diffusion weighted images, which give rise to regression tasks on the manifold GL ( n )/ O ( n ) for diffusion tensor images (DTI) and the Hilbert unit sphere for orientation distribution functions (ODF) from high angular resolution acquisition. The companion open-source code is available on nitrc.org/projects/riem_mglm.- Published
- 2014
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.