Despite the powerful capabilities of GNN-based drug screening model in predicting target drug properties, the black-box nature of these models poses a challenge for practical application, particularly in a field as critical as drug development where understanding and trust in AI-driven decisions are important. To address the interpretability issues associated with GNN-based virtual drug screening, we introduce XplainScreen: a unified explanation framework designed to evaluate various explanation methods for GNN-based models. XplainScreen offers a user-friendly, web-based interactive platform that allows for the selection of specific GNN-based drug screening models and multiple cutting-edge explainable AI methods. It supports both qualitative assessments (through visualization and generative text descriptions) and quantitative evaluations of these methods, utilizing drug molecules in SMILES format. This demonstration showcases the utility of XplainScreen through a user study with pharmacological researchers focused on virtual screening tasks based on toxicity, highlighting the framework's potential to enhance the integrity and trustworthiness of AI-driven virtual drug screening. A video demo of XplainScreen is available at https://youtu.be/Q4yobrTLKec, and the source code can be accessed at https://github.com/GeonHeeAhn/XplainScreen.

In this study, we propose a solution to estimating system responses to external forcings or perturbations. We utilize the Fluctuation-Dissipation Theorem (FDT) from statistical physics to extract knowledge using an AI model that can rapidly produce scenarios for different external forcings by leveraging FDT and analyzing a large dataset from Earth System Models. Our model, AiBEDO, accurately captures the complex effects of radiation perturbations on global and regional surface climate, enabling faster exploration of the impacts of spatially-heterogenous climate forcings. We demonstrate its effectiveness by applying AiBEDO to Marine Cloud Brightening, a climate intervention technique, aiming to optimize cloud brightening patterns for regional climate targets and prevent climate tipping points. Our approach has broader applicability to other scientific disciplines with computationally demanding simulation models. Source code of AiBEDO framework is made available at https://github.com/kramea/cikm_aibedo. A sample dataset is made available at https://doi.org/10.5281/zenodo.7597027. Additional data available upon request.

Complex computational models are often designed to simulate real-world physical phenomena in many scientific disciplines. However, these simulation models tend to be computationally very expensive and involve a large number of simulation input parameters, which need to be analyzed and properly calibrated before the models can be applied for real scientific studies. We propose a visual analysis system to facilitate interactive exploratory analysis of high-dimensional input parameter space for a complex yeast cell polarization simulation. The proposed system can assist the computational biologists, who designed the simulation model, to visually calibrate the input parameters by modifying the parameter values and immediately visualizing the predicted simulation outcome without having the need to run the original expensive simulation for every instance. Our proposed visual analysis system is driven by a trained neural network-based surrogate model as the backend analysis framework. In this work, we demonstrate the advantage of using neural networks as surrogate models for visual analysis by incorporating some of the recent advances in the field of uncertainty quantification, interpretability and explainability of neural network-based models. We utilize the trained network to perform interactive parameter sensitivity analysis of the original simulation as well as recommend optimal parameter configurations using the activation maximization framework of neural networks. We also facilitate detail analysis of the trained network to extract useful insights about the simulation model, learned by the network, during the training process. We performed two case studies, and discovered multiple new parameter configurations, which can trigger high cell polarization results in the original simulation model. We evaluated our results by comparing with the original simulation model outcomes as well as the findings from previous parameter analysis performed by our experts.

Convolutional neural networks are one of the most important and widely used constructs in natural language processing and AI in general. In many applications, they have achieved state-of-the-art performance, with training time faster than the other alternatives. However, due to their limited interpretability, they are less favored by practitioners over attention-based models, like RNNs and self-attention (Transformers), which can be visualized and interpreted more intuitively by analyzing the attention-weight heat-maps. In this work, we present a visualization technique that can be used to understand the inner workings of text-based CNN models. We also show how this method can be used to generate adversarial examples and learn the shortcomings of the training data.

CoDDA (Copula-based Distribution Driven Analysis) is a flexible framework for large-scale multivariate datasets. A common strategy to deal with large-scale scientific simulation data is to partition the simulation domain and create statistical data summaries. Instead of storing the high-resolution raw data from the simulation, storing the compact statistical data summaries results in reduced storage overhead and alleviated I/O bottleneck. Such summaries, often represented in the form of statistical probability distributions, can serve various post-hoc analysis and visualization tasks. However, for multivariate simulation data using standard multivariate distributions for creating data summaries is not feasible. They are either storage inefficient or are computationally expensive to be estimated in simulation time (in situ) for large number of variables. In this work, using copula functions, we propose a flexible multivariate distribution-based data modeling and analysis framework that offers significant data reduction and can be used in an in situ environment. The framework also facilitates in storing the associated spatial information along with the multivariate distributions in an efficient representation. Using the proposed multivariate data summaries, we perform various multivariate post-hoc analyses like query-driven visualization and sampling-based visualization. We evaluate our proposed method on multiple real-world multivariate scientific datasets. To demonstrate the efficacy of our framework in an in situ environment, we apply it on a large-scale flow simulation.

Over the last decade, ensemble visualization has witnessed a significant development due to the wide availability of ensemble data, and the increasing visualization needs from a variety of disciplines. From the data analysis point of view, it can be observed that many ensemble visualization works focus on the same facet of ensemble data, use similar data aggregation or uncertainty modeling methods. However, the lack of reflections on those essential commonalities and a systematic overview of those works prevents visualization researchers from effectively identifying new or unsolved problems and planning for further developments. In this paper, we take a holistic perspective and provide a survey of ensemble visualization. Specifically, we study ensemble visualization works in the recent decade, and categorize them from two perspectives: (1) their proposed visualization techniques; and (2) their involved analytic tasks. For the first perspective, we focus on elaborating how conventional visualization techniques (e.g., surface, volume visualization techniques) have been adapted to ensemble data; for the second perspective, we emphasize how analytic tasks (e.g., comparison, clustering) have been performed differently for ensemble data. From the study of ensemble visualization literature, we have also identified several research trends, as well as some future research opportunities.

Distributions are often used to model uncertainty in many scientific datasets. To preserve the correlation among the spatially sampled grid locations in the dataset, various standard multivariate distribution models have been proposed in visualization literature. These models treat each grid location as a univariate random variable which models the uncertainty at that location. Standard multivariate distributions (both parametric and nonparametric) assume that all the univariate marginals are of the same type/family of distribution. But in reality, different grid locations show different statistical behavior which may not be modeled best by the same type of distribution. In this paper, we propose a new multivariate uncertainty modeling strategy to address the needs of uncertainty modeling in scientific datasets. Our proposed method is based on a statistically sound multivariate technique called Copula, which makes it possible to separate the process of estimating the univariate marginals and the process of modeling dependency, unlike the standard multivariate distributions. The modeling flexibility offered by our proposed method makes it possible to design distribution fields which can have different types of distribution (Gaussian, Histogram, KDE etc.) at the grid locations, while maintaining the correlation structure at the same time. Depending on the results of various standard statistical tests, we can choose an optimal distribution representation at each location, resulting in a more cost efficient modeling without significantly sacrificing on the analysis quality. To demonstrate the efficacy of our proposed modeling strategy, we extract and visualize uncertain features like isocontours and vortices in various real world datasets. We also study various modeling criterion to help users in the task of univariate model selection.

Visualizing the similarities and differences among an ensemble of isosurfaces is a challenging problem mainly because the isosurfaces cannot be displayed together at the same time. For ensemble of isosurfaces, visualizing these spatial differences among the surfaces is essential to get useful insights as to how the individual ensemble simulations affect different isosurfaces. We propose a scheme to visualize the spatial variations of isosurfaces with respect to statistically significant isosurfaces within the ensemble. Understanding such variations among ensemble of isosurfaces at different spatial regions is helpful in analyzing the influence of different ensemble runs over the spatial domain. In this regard, we propose an isosurface-entropy based clustering scheme to divide the spatial domain into regions of high and low isosurface variation. We demonstrate the efficacy of our method by successfully applying it on real-world ensemble data sets from ocean simulation experiments and weather forecasts.