Research in ML4VIS investigates how to use machine learning (ML) techniques to generate visualizations, and the field is rapidly growing with high societal impact. However, as with any computational pipeline that employs ML processes, ML4VIS approaches are susceptible to a range of ML-specific adversarial attacks. These attacks can manipulate visualization generations, causing analysts to be tricked and their judgments to be impaired. Due to a lack of synthesis from both visualization and ML perspectives, this security aspect is largely overlooked by the current ML4VIS literature. To bridge this gap, we investigate the potential vulnerabilities of ML-aided visualizations from adversarial attacks using a holistic lens of both visualization and ML perspectives. We first identify the attack surface (i.e., attack entry points) that is unique in ML-aided visualizations. We then exemplify five different adversarial attacks. These examples highlight the range of possible attacks when considering the attack surface and multiple different adversary capabilities. Our results show that adversaries can induce various attacks, such as creating arbitrary and deceptive visualizations, by systematically identifying input attributes that are influential in ML inferences. Based on our observations of the attack surface characteristics and the attack examples, we underline the importance of comprehensive studies of security issues and defense mechanisms as a call of urgency for the ML4VIS community.

Synchronous data-driven storytelling with network visualiza-tions presents significant challenges due to the complexity of real-time manipulation of network components. While existingresearch addresses asynchronous scenarios, there is a lack of effective tools for live presentations. To address this gap, we de-veloped TangibleNet, a projector-based AR prototype that allows presenters to interact with node-link diagrams using double-sided magnets during live presentations. The design process was informed by interviews with professionals experienced insynchronous data storytelling and workshops with 14 HCI/VISresearchers. Insights from the interviews helped identify key design considerations for integrating physical objects as interac-tive tools in presentation contexts. The workshops contributed to the development of a design space mapping user actions to interaction commands for node-link diagrams. Evaluation with 12 participants confirmed that TangibleNet supports intuitive in-teractions and enhances presenter autonomy, demonstrating its effectiveness for synchronous network-based data storytelling.

Multi-class scatterplots are essential for visually comparing data, such as examining class distributions in dimensionality reduction and evaluating classification models. Visual class separation (VCS) measures quantify human perception but are largely derived from and evaluated with datasets reflecting limited types of scatterplot features (e.g., data distribution, similar class densities). Quantitatively identifying which scatterplot features are influential to VCS tasks can enable more robust guidance for future measures. We analyze the alignment between VCS measures and people's perceptions of class separation through a crowdsourced study using 70 scatterplot features relevant to class separation. To cover a wide range of scatterplot features, we generated a set of multi-class scatterplots from 6,947 real-world datasets. Our results highlight that multiple combinations of features are needed to best explain VCS. From our analysis, we develop a composite feature model that identifies key scatterplot features for measuring VCS task performance.

Dimensionality reduction (DR) techniques are essential for visually analyzing high-dimensional data. However, visual analytics using DR often face unreliability, stemming from factors such as inherent distortions in DR projections. This unreliability can lead to analytic insights that misrepresent the underlying data, potentially resulting in misguided decisions. To tackle these reliability challenges, we review 133 papers that address the unreliability of visual analytics using DR. Through this review, we contribute (1) a workflow model that describes the interaction between analysts and machines in visual analytics using DR, and (2) a taxonomy that identifies where and why reliability issues arise within the workflow, along with existing solutions for addressing them. Our review reveals ongoing challenges in the field, whose significance and urgency are validated by five expert researchers. This review also finds that the current research landscape is skewed toward developing new DR techniques rather than their interpretation or evaluation, where we discuss how the HCI community can contribute to broadening this focus.

One advancing machine-learning-based analysis approach for multivariate time-series data is representing data as a third-order tensor and then applying dimensionality reduction (DR) methods. In this work, we introduce a visual analytics method that employs multiple interactive DR methods to support both extraction and interpretation of latent patterns of multivariate time-series data. Our method first allows analysts to select an analysis focus from three axes: instance, variable, and time axes. Then, the method applies a multi-step DR method to produce a 2D scatterplot that depicts latent patterns of the selected axis's elements (e.g., time points). Afterward, the analysts interactively investigate data groups that appeared in the plot with a DR method designed for comparative analysis. The method can be further applied iteratively to perform more precise and detailed analyses. We implement a prototype system and demonstrate the effectiveness of our method by analyzing supercomputer log data.

Although many dimensionality reduction (DR) techniques employ stochastic methods for computational efficiency, such as negative sampling or stochastic gradient descent, their impact on the projection has been underexplored. In this work, we investigate how such stochasticity affects the stability of projections and present a novel DR technique, GhostUMAP, to measure the pointwise instability of projections. Our idea is to introduce clones of data points, "ghosts", into UMAP’s layout optimization process. Ghosts are designed to be completely passive: they do not affect any others but are influenced by attractive and repulsive forces from the original data points. After a single optimization run, GhostUMAP can capture the projection instability of data points by measuring the variance with the projected positions of their ghosts. We also present a successive halving technique to reduce the computation of GhostUMAP. Our results suggest that Ghost-UMAP can reveal unstable data points with a reasonable computational overhead.

Flagship-class high-performance computing (HPC) systems, also known as supercomputers, are large, complex systems that require particular attention for continuous and long-term stable operations. The K computer was a Japanese flagship-class supercomputer ranked as the fastest supercomputer in the Top500 ranking when it first appeared. It was composed of more than eighty thousand compute nodes and consumed more than 12 MW when running the LINPACK benchmark for the Top500 submission. A combined power substation, with a natural gas co-generation system (CGS), was used for the power supply, and also a large air/water cooling facility was used to extract the massive heat generated from this HPC system. During the years of its regular operation, a large log dataset has been generated from the K computer system and its facility, and several visual analytics systems have been developed to better understand the K computer's behavior during the operation as well as the probable correlation of operational temperature with the critical hardware failures. In this paper, we will reflect on these visual analytics systems, mainly developed by graduate students, intended to be used by different types of end users on the HPC site. In addition, we will discuss the importance of collaborative development involving the end users, and also the importance of technical people in the middle for assisting in the deployment and possible continuation of the developed systems.

The size and complexity of supercomputer systems and their power and cooling facilities have continuously increased, thus posing additional challenge for long-term and stable operation. Supercomputers are shared computational resources and usually operate with different computational workloads at different locations (space) and timings (time). Better understanding of the supercomputer systems heat generation and cooling behavior is highly desired from the facility operational side for decision making and optimization planning. In this work, we present a dimensionality reduction-based visual analytics method for time-series log data, from supercomputer system and its facility, to capture characteristic spatio-temporal features and behaviors during the operation.

Identifying unique characteristics in a network through comparison with another network is an essential network analysis task. For example, with networks of protein interactions obtained from normal and cancer tissues, we can discover unique types of interactions in cancer tissues. This analysis task could be greatly assisted by contrastive learning, which is an emerging analysis approach to discover salient patterns in one dataset relative to another. However, existing contrastive learning methods cannot be directly applied to networks as they are designed only for high-dimensional data analysis. To address this problem, we introduce a new analysis approach called contrastive network representation learning (cNRL). By integrating two machine learning schemes, network representation learning and contrastive learning, cNRL enables embedding of network nodes into a low-dimensional representation that reveals the uniqueness of one network compared to another. Within this approach, we also design a method, named i-cNRL, which offers interpretability in the learned results, allowing for understanding which specific patterns are only found in one network. We demonstrate the effectiveness of i-cNRL for network comparison with multiple network models and real-world datasets. Furthermore, we compare i-cNRL and other potential cNRL algorithm designs through quantitative and qualitative evaluations.