Author name cluster

Dongwoo Kim

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

12 papers

2 author rows

AAAI Conference 2026 Conference Paper

Posterior Label Smoothing for Node Classification

Jaeseung Heo
MoonJeong Park
Dongwoo Kim

Label smoothing is a widely studied regularization technique in machine learning. However, its potential for node classification in graph-structured data, spanning homophilic to heterophilic graphs, remains largely unexplored. We introduce posterior label smoothing, a novel method for transductive node classification that derives soft labels from a posterior distribution conditioned on neighborhood labels. The likelihood and prior distributions are estimated from the global statistics of the graph structure, allowing our approach to adapt naturally to various graph properties. We evaluate our method on 10 benchmark datasets using eight baseline models, demonstrating consistent improvements in classification accuracy. The following analysis demonstrates that soft labels mitigate overfitting during training, leading to better generalization performance, and that pseudo-labeling effectively refines the global label statistics of the graph.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

High-order Equivariant Flow Matching for Density Functional Theory Hamiltonian Prediction

Seongsu Kim
Nayoung Kim
Dongwoo Kim
Sungsoo Ahn

Density functional theory (DFT) is a fundamental method for simulating quantum chemical properties, but it remains expensive due to the iterative self-consistent field (SCF) process required to solve the Kohn–Sham equations. Recently, deep learning methods are gaining attention as a way to bypass this step by directly predicting the Hamiltonian. However, they rely on deterministic regression and do not consider the highly structured nature of Hamiltonians. In this work, we propose QHFlow, a high-order equivariant flow matching framework that generates Hamiltonian matrices conditioned on molecular geometry. Flow matching models continuous-time trajectories between simple priors and complex targets, learning the structured distributions over Hamiltonians instead of direct regression. To further incorporate symmetry, we use a neural architecture that predicts SE(3)-equivariant vector fields, improving accuracy and generalization across diverse geometries. To further enhance physical fidelity, we additionally introduce a fine-tuning scheme to align predicted orbital energies with the target. QHFlow achieves state-of-the-art performance, reducing Hamiltonian error by 71% on MD17 and 53% on QH9. Moreover, we further show that QHFlow accelerates the DFT process without trading off the solution quality when initializing SCF iterations with the predicted Hamiltonian, significantly reducing the number of iterations and runtime.

PDF Details

NeurIPS Conference 2025 Conference Paper

Influence Functions for Edge Edits in Non-Convex Graph Neural Networks

Jaeseung Heo
Kyeongheung Yun
Seokwon Yoon
MoonJeong Park
Jungseul Ok
Dongwoo Kim

Understanding how individual edges influence the behavior of graph neural networks (GNNs) is essential for improving their interpretability and robustness. Graph influence functions have emerged as promising tools to efficiently estimate the effects of edge deletions without retraining. However, existing influence prediction methods rely on strict convexity assumptions, exclusively consider the influence of edge deletions while disregarding edge insertions, and fail to capture changes in message propagation caused by these modifications. In this work, we propose a proximal Bregman response function specifically tailored for GNNs, relaxing the convexity requirement and enabling accurate influence prediction for standard neural network architectures. Furthermore, our method explicitly accounts for message propagation effects and extends influence prediction to both edge deletions and insertions in a principled way. Experiments with real-world datasets demonstrate accurate influence predictions for different characteristics of GNNs. We further demonstrate that the influence function is versatile in applications such as graph rewiring and adversarial attacks.

PDF Details

IJCAI Conference 2024 Conference Paper

EPIC: Graph Augmentation with Edit Path Interpolation via Learnable Cost

Jaeseung Heo
Seungbeom Lee
Sungsoo Ahn
Dongwoo Kim

Data augmentation plays a critical role in improving model performance across various domains, but it becomes challenging with graph data due to their complex and irregular structure. To address this issue, we propose EPIC (Edit Path Interpolation via learnable Cost), a novel interpolation-based method for augmenting graph datasets. To interpolate between two graphs lying in an irregular domain, EPIC leverages the concept of graph edit distance, constructing an edit path that represents the transformation process between two graphs via edit operations. Moreover, our method introduces a context-sensitive cost model that accounts for the importance of specific edit operations formulated through a learning framework. This allows for a more nuanced transformation process, where the edit distance is not merely count-based but reflects meaningful graph attributes. With randomly sampled graphs from the edit path, we enrich the training set to enhance the generalization capability of classification models. Experimental evaluations across several benchmark datasets demonstrate that our approach outperforms existing augmentation techniques in many tasks.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Feature Unlearning for Pre-trained GANs and VAEs

Saemi Moon
Seunghyuk Cho
Dongwoo Kim

We tackle the problem of feature unlearning from a pre-trained image generative model: GANs and VAEs. Unlike a common unlearning task where an unlearning target is a subset of the training set, we aim to unlearn a specific feature, such as hairstyle from facial images, from the pre-trained generative models. As the target feature is only presented in a local region of an image, unlearning the entire image from the pre-trained model may result in losing other details in the remaining region of the image. To specify which features to unlearn, we collect randomly generated images that contain the target features. We then identify a latent representation corresponding to the target feature and then use the representation to fine-tune the pre-trained model. Through experiments on MNIST, CelebA, and FFHQ datasets, we show that target features are successfully removed while keeping the fidelity of the original models. Further experiments with an adversarial attack show that the unlearned model is more robust under the presence of malicious parties.

PDF Details DOI

AAAI Conference 2023 Conference Paper

Anonymization for Skeleton Action Recognition

Saemi Moon
Myeonghyeon Kim
Zhenyue Qin
Yang Liu
Dongwoo Kim

Skeleton-based action recognition attracts practitioners and researchers due to the lightweight, compact nature of datasets. Compared with RGB-video-based action recognition, skeleton-based action recognition is a safer way to protect the privacy of subjects while having competitive recognition performance. However, due to improvements in skeleton recognition algorithms as well as motion and depth sensors, more details of motion characteristics can be preserved in the skeleton dataset, leading to potential privacy leakage. We first train classifiers to categorize private information from skeleton trajectories to investigate the potential privacy leakage from skeleton datasets. Our preliminary experiments show that the gender classifier achieves 87% accuracy on average, and the re-identification classifier achieves 80% accuracy on average with three baseline models: Shift-GCN, MS-G3D, and 2s-AGCN. We propose an anonymization framework based on adversarial learning to protect potential privacy leakage from the skeleton dataset. Experimental results show that an anonymized dataset can reduce the risk of privacy leakage while having marginal effects on action recognition performance even with simple anonymizer architectures. The code used in our experiments is available at https://github.com/ml-postech/Skeleton-anonymization/

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Hyperbolic VAE via Latent Gaussian Distributions

Seunghyuk Cho
Juyong Lee
Dongwoo Kim

We propose a Gaussian manifold variational auto-encoder (GM-VAE) whose latent space consists of a set of Gaussian distributions. It is known that the set of the univariate Gaussian distributions with the Fisher information metric form a hyperbolic space, which we call a Gaussian manifold. To learn the VAE endowed with the Gaussian manifolds, we propose a pseudo-Gaussian manifold normal distribution based on the Kullback-Leibler divergence, a local approximation of the squared Fisher-Rao distance, to define a density over the latent space. We demonstrate the efficacy of GM-VAE on two different tasks: density estimation of image datasets and state representation learning for model-based reinforcement learning. GM-VAE outperforms the other variants of hyperbolic- and Euclidean-VAEs on density estimation tasks and shows competitive performance in model-based reinforcement learning. We observe that our model provides strong numerical stability, addressing a common limitation reported in previous hyperbolic-VAEs. The implementation is available at https: //github. com/ml-postech/GM-VAE.

PDF Details

EAAI Journal 2023 Journal Article

Photo-realistic 3D model based accurate visual positioning system for large-scale indoor spaces

Janghun Hyeon
Bumchul Jang
Hyunga Choi
Joohyung Kim
Dongwoo Kim
Nakju Doh

This study presents a novel and reliable visual positioning system (VPS), KR-Net, for kidnap recovery tasks, which predicts an accurate position when a robot is first initiated. KR-Net is based on a hierarchical visual localization method and demonstrates significant robustness in large-scale indoor environments. The proposed VPS utilizes a photo-realistic 3D model to generate a dense database of any camera pose and incorporates a novel global descriptor for indoor spaces, i-GeM, that outperforms existing methods in terms of robustness. Additionally, the proposed combinatorial pooling approach overcomes the limitations of previous single image-based predictions in large-scale indoor environments, allowing for accurate discrimination between similar locations. Extensive evaluations were performed on six large-scale indoor datasets to demonstrate the contributions of each component. To the best of our knowledge, KR-Net is the first system to estimate wake-up positions with a near 100% confidence level within a 1. 0 m distance error threshold.

Details DOI

AAAI Conference 2023 Conference Paper

Restructuring Graph for Higher Homophily via Adaptive Spectral Clustering

Shouheng Li
Dongwoo Kim
Qing Wang

While a growing body of literature has been studying new Graph Neural Networks (GNNs) that work on both homophilic and heterophilic graphs, little has been done on adapting classical GNNs to less-homophilic graphs. Although the ability to handle less-homophilic graphs is restricted, classical GNNs still stand out in several nice properties such as efficiency, simplicity, and explainability. In this work, we propose a novel graph restructuring method that can be integrated into any type of GNNs, including classical GNNs, to leverage the benefits of existing GNNs while alleviating their limitations. Our contribution is threefold: a) learning the weight of pseudo-eigenvectors for an adaptive spectral clustering that aligns well with known node labels, b) proposing a new density-aware homophilic metric that is robust to label imbalance, and c) reconstructing the adjacency matrix based on the result of adaptive spectral clustering to maximize the homophilic scores. The experimental results show that our graph restructuring method can significantly boost the performance of six classical GNNs by an average of 25% on less-homophilic graphs. The boosted performance is comparable to state-of-the-art methods.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

A Rotated Hyperbolic Wrapped Normal Distribution for Hierarchical Representation Learning

Seunghyuk Cho
Juyong Lee
Jaesik Park
Dongwoo Kim

We present a rotated hyperbolic wrapped normal distribution (RoWN), a simple yet effective alteration of a hyperbolic wrapped normal distribution (HWN). The HWN expands the domain of probabilistic modeling from Euclidean to hyperbolic space, where a tree can be embedded with arbitrary low distortion in theory. In this work, we analyze the geometric properties of the diagonal HWN, a standard choice of distribution in probabilistic modeling. The analysis shows that the distribution is inappropriate to represent the data points at the same hierarchy level through their angular distance with the same norm in the Poincar\'e disk model. We then empirically verify the presence of limitations of HWN, and show how RoWN, the proposed distribution, can alleviate the limitations on various hierarchical datasets, including noisy synthetic binary tree, WordNet, and Atari 2600 Breakout. The code is available at https: //github. com/ml-postech/RoWN.

PDF Details

IROS Conference 2020 Conference Paper

KR-Net: A Dependable Visual Kidnap Recovery Network for Indoor Spaces

Janghun Hyeon
Dongwoo Kim
Bumchul Jang
Hyunga Choi
Dong Hoon Yi
Kyungho Yoo
Jeongae Choi
Nakju Lett Doh

In this paper, we propose a dependable visual kidnap recovery (KR) framework that pinpoints a unique pose in a given 3D map when a device is turned on. For this framework, we first develop indoor-GeM (i-GeM), which is an extension of GeM [1] but considerably more robust than other global descriptors [2]-[4], including GeM itself. Then, we propose a convolutional neural network (CNN)-based system called KR-Net, which is based on a coarse-to-fine paradigm as in [5] and [6]. To our knowledge, KR-Net is the first network that can pinpoint a wake-up pose with a confidence level near 100% within a 1. 0 m translational error boundary. This dependable success rate is enabled not only by i-GeM, but also by a combinatorial pooling approach that uses multiple images around the wake-up spot, whereas previous implementations [5], [6] were constrained to a single image. Experiments were conducted in two challenging datasets: a large-scale (12, 557 m 2 ) area with frequent featureless or repetitive places and a place with significant view changes due to a one-year gap between prior modeling and query acquisition. Given 59 test query sets (eight images per pose), KR-Net successfully found all wake-up poses, with average and maximum errors of 0. 246 m and 0. 983 m, respectively.

Details

IJCAI Conference 2013 Conference Paper

Context-Dependent Conceptualization

Dongwoo Kim
Haixun Wang
Alice Oh

Conceptualization seeks to map a short text (i. e. , a word or a phrase) to a set of concepts as a mechanism of understanding text. Most of prior research in conceptualization uses human-crafted knowledge bases that map instances to concepts. Such approaches to conceptualization have the limitation that the mappings are not context sensitive. To overcome this limitation, we propose a framework in which we harness the power of a probabilistic topic model which inherently captures the semantic relations between words. By combining latent Dirichlet allocation, a widely used topic model with Probase, a large-scale probabilistic knowledge base, we develop a corpus-based framework for context-dependent conceptualization. Through this simple but powerful framework, we improve conceptualization and enable a wide range of applications that rely on semantic understanding of short texts, including frame element prediction, word similarity in context, ad-query similarity, and query similarity.

PDF Details DOI