Author name cluster

Kai Hwang

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.

5 papers

1 author row

AAAI Conference 2025 Conference Paper

Unsupervised Anomaly Detection for Tabular Data Using Deep Noise Evaluation

Wei Dai
Kai Hwang
Jicong Fan

Unsupervised anomaly detection (UAD) plays an important role in modern data analytics and it is crucial to provide simple yet effective and guaranteed UAD algorithms for real applications. In this paper, we present a novel UAD method for tabular data by evaluating how much noise is in the data. Specifically, we propose to learn a deep neural network from the clean (normal) training dataset and a noisy dataset, where the latter is generated by adding highly diverse noises to the clean data. The neural network can learn a reliable decision boundary between normal data and anomalous data when the diversity of the generated noisy data is sufficiently high so that the hard abnormal samples lie in the noisy region. Importantly, we provide theoretical guarantees, proving that the proposed method can detect anomalous data successfully, although the method does not utilize any real anomalous data in the training stage. Extensive experiments through more than 60 benchmark datasets demonstrate the effectiveness of the proposed method in comparison to 12 baselines of UAD. Our method obtains a 92.27% AUC score and a 1.68 ranking score on average. Moreover, compared to the state-of-the-art UAD methods, our method is easier to implement.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

Rethinking the Effectiveness of Graph Classification Datasets in Benchmarks for Assessing GNNs

Zhengdao Li
Yong Cao
Kefan Shuai
Yiming Miao
Kai Hwang

Graph classification benchmarks, vital for assessing and developing graph neural network (GNN) models, have recently been scrutinized, as simple methods like MLPs have demonstrated comparable performance. This leads to an important question: Do these benchmarks effectively distinguish the advancements of GNNs over other methodologies? If so, how do we quantitatively measure this effectiveness? In response, we first propose an empirical protocol based on a fair benchmarking framework to investigate the performance discrepancy between simple methods and GNNs. We further propose a novel metric to quantify the dataset effectiveness by considering both dataset complexity and model performance. To the best of our knowledge, our work is the first to thoroughly study and provide an explicit definition for dataset effectiveness in the graph learning area. Through testing across 16 real-world datasets, we found our metric to align with existing studies and intuitive assumptions. Finally, we explore the causes behind the low effectiveness of certain datasets by investigating the correlation between intrinsic graph properties and class labels, and we developed a novel technique supporting the correlation-controllable synthetic dataset generation. Our findings shed light on the current understanding of benchmark datasets, and our new platform could fuel the future evolution of graph classification benchmarks.

PDF Details DOI

YNIMG Journal 2016 Journal Article

Frontal preparatory neural oscillations associated with cognitive control: A developmental study comparing young adults and adolescents

Kai Hwang
Avniel S. Ghuman
Dara S. Manoach
Stephanie R. Jones
Beatriz Luna

Functional magnetic resonance imaging (fMRI) studies suggest that age-related changes in the frontal cortex may underlie developmental improvements in cognitive control. In the present study we used magnetoencephalography (MEG) to identify frontal oscillatory neurodynamics that support age-related improvements in cognitive control during adolescence. We characterized the differences in neural oscillations in adolescents and adults during the preparation to suppress a prepotent saccade (antisaccade trials—AS) compared to preparing to generate a more automatic saccade (prosaccade trials—PS). We found that for adults, AS were associated with increased beta-band (16–38Hz) power in the dorsal lateral prefrontal cortex (DLPFC), enhanced alpha- to low beta-band (10–18Hz) power in the frontal eye field (FEF) that predicted performance, and increased cross-frequency alpha-beta (10–26Hz) amplitude coupling between the DLPFC and the FEF. Developmental comparisons between adults and adolescents revealed similar engagement of DLPFC beta-band power but weaker FEF alpha-band power, and lower cross-frequency coupling between the DLPFC and the FEF in adolescents. These results suggest that lateral prefrontal neural activity associated with cognitive control is adult-like by adolescence; the development of cognitive control from adolescence to adulthood is instead associated with increases in frontal connectivity and strengthening of inhibition signaling for suppressing task-incompatible processes.

Details DOI

YNIMG Journal 2013 Journal Article

The nuisance of nuisance regression: Spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity

Michael N. Hallquist
Kai Hwang
Beatriz Luna

Recent resting-state functional connectivity fMRI (RS-fcMRI) research has demonstrated that head motion during fMRI acquisition systematically influences connectivity estimates despite bandpass filtering and nuisance regression, which are intended to reduce such nuisance variability. We provide evidence that the effects of head motion and other nuisance signals are poorly controlled when the fMRI time series are bandpass-filtered but the regressors are unfiltered, resulting in the inadvertent reintroduction of nuisance-related variation into frequencies previously suppressed by the bandpass filter, as well as suboptimal correction for noise signals in the frequencies of interest. This is important because many RS-fcMRI studies, including some focusing on motion-related artifacts, have applied this approach. In two cohorts of individuals (n =117 and 22) who completed resting-state fMRI scans, we found that the bandpass–regress approach consistently overestimated functional connectivity across the brain, typically on the order of r =. 10–. 35, relative to a simultaneous bandpass filtering and nuisance regression approach. Inflated correlations under the bandpass–regress approach were associated with head motion and cardiac artifacts. Furthermore, distance-related differences in the association of head motion and connectivity estimates were much weaker for the simultaneous filtering approach. We recommend that future RS-fcMRI studies ensure that the frequencies of nuisance regressors and fMRI data match prior to nuisance regression, and we advocate a simultaneous bandpass filtering and nuisance regression strategy that better controls nuisance-related variability.

Details DOI

YNIMG Journal 2009 Journal Article

Category-specific activations during word generation reflect experiential sensorimotor modalities

Kai Hwang
Erica D. Palmer
Surina Basho
Jonathan R. Zadra
Ralph-Axel Müller

According to the sensorimotor theory of lexicosemantic organization, semantic representations are neurally distributed and anatomically linked to category-specific sensory areas. Previous functional neuroimaging studies have demonstrated category specificity in lexicosemantic representations. However, little evidence is available from word generation paradigms, which provide access to semantic representations while minimizing confounds resulting from low-level perceptual features of stimulus presentation. In this study, 13 healthy young adults underwent fMRI scanning while performing a word generation task, generating exemplars to nine different semantic categories. Each semantic category was assigned to one of three superordinate category types, based upon sensorimotor modalities (visual, motor, somatosensory) presumed to predominate in lexical acquisition. For word generation overall, robust activation was seen in left inferior frontal cortex. Analyses by sensorimotor modality categories yielded activations in brain regions related to perceptual and motor processing: Visual categories activated extrastriate cortex, motor categories activated the intraparietal sulcus and posterior middle temporal cortex, and somatosensory categories activated postcentral and inferior parietal regions. Our results are consistent with the sensorimotor theory, according to which lexicosemantic representations are distributed across brain regions participating in sensorimotor processing associated with the experiential components of lexicosemantic acquisition.

Details DOI