arXiv subjects

Wooseok Jung

Publications and source records attributed to Wooseok Jung.

3 recordsLinked to original sources

arXiv2026

An agent-based model of the formation and evolution of common ground

The existence of a communal common ground is vital for collective action and coordination in a population, but the micro-level cognitive and social processes that lead to the formation and evolution of common ground at the macro-level are undertheorised and have not been rigorously explored. In this work, we adopt a formal approach and develop an agent-based model that describes repeated grounding attempts between agents interacting on a network, with an explicit distinction between a sender agent and a receiver agent during an interaction involving sharing information. Several key novel features enable us to capture a range of different interaction contexts: we allow for the interaction to result in either acceptance or rejection, the receiver's response may be lost to the sender, and the sender can interpret this lack of response as either acceptance or rejection (or even something in between). A campaign of Monte Carlo simulations reveals how different interaction contexts, as well as the available information for sharing, result in different emergent phenomena, such as a global communal common ground, fragmentation into multiple clusters of differing common ground, and even the total loss of any shared common ground. This work highlights the potential for using mathematical models to study micro-macro links in cultural dynamics, including identifying ways to facilitate interactions to foster the emergence of a global communal common ground.

physics.soc-ph↗

arXiv2021

Persistent homology method to detect block structures in weighted networks

Unravelling the block structure of a network is critical for studying macroscopic features and community-level dynamics. The weighted stochastic block model (WSBM), a variation of the traditional stochastic block model, is designed for weighted networks, but it is not always optimal. We introduce a novel topological method to study the block structure of weighted networks by comparing their persistence diagrams. We found persistence diagrams of networks with different block structures show distinct features, sufficient to distinguish. Moreover, the overall characteristics are preserved even with more stochastic examples or modified hyperparameters. Finally, when random graphs whose latent block structure is unknown are tested, results from persistence diagram analysis are consistent with their weighted stochastic block model. Although this topological method cannot completely replace the original WSBM method for some reasons, it is worth to be investigated further.

math.DS↗

arXiv2021

From augmented microscopy to the topological transformer: a new approach in cell image analysis for Alzheimer's research

Cell image analysis is crucial in Alzheimer's research to detect the presence of A$β$ protein inhibiting cell function. Deep learning speeds up the process by making only low-level data sufficient for fruitful inspection. We first found Unet is most suitable in augmented microscopy by comparing performance in multi-class semantics segmentation. We develop the augmented microscopy method to capture nuclei in a brightfield image and the transformer using Unet model to convert an input image into a sequence of topological information. The performance regarding Intersection-over-Union is consistent concerning the choice of image preprocessing and ground-truth generation. Training model with data of a specific cell type demonstrates transfer learning applies to some extent. The topological transformer aims to extract persistence silhouettes or landscape signatures containing geometric information of a given image of cells. This feature extraction facilitates studying an image as a collection of one-dimensional data, substantially reducing computational costs. Using the transformer, we attempt grouping cell images by their cell type relying solely on topological features. Performances of the transformers followed by SVM, XGBoost, LGBM, and simple convolutional neural network classifiers are inferior to the conventional image classification. However, since this research initiates a new perspective in biomedical research by combining deep learning and topology for image analysis, we speculate follow-up investigation will reinforce our genuine regime.

eess.IV↗