VISITOR: Visual Interactive State Sequence Exploration for Reinforcement Learning

Understanding the behavior of deep reinforcement learning agents is a crucial requirement throughout their development. Existing work has addressed the identification of observable behavioral patterns in state sequences or analysis of isolated internal representations; however, the overall decision-making of deep-learning RL agents remains opaque. To tackle this, we present VISITOR, a visual analytics system enabling the analysis of entire state sequences, the diagnosis of singular predictions, and the comparison between agents. A sequence embedding view enables the multiscale analysis of state sequences, utilizing custom embedding techniques for a stable spatialization of the observations and internal states. We provide multiple layers: (1) a state space embedding, highlighting different groups of states inside the state-action sequences, (2) a trajectory view, emphasizing decision points, (3) a network activation mapping, visualizing the relationship between observations and network activations, (4) a transition embedding, enabling the analysis of state-to-state transitions. The embedding view is accompanied by an interactive reward view that captures the temporal development of metrics, which can be linked directly to states in the embedding. Lastly, a model list allows for the quick comparison of models across multiple metrics. Annotations can be exported to communicate results to different audiences. Our two-stage evaluation with eight experts confirms the effectiveness in identifying states of interest, comparing the quality of policies, and reasoning about the internal decision-making processes.     

Visual Inspection of Multivariate Graphs

Most graph visualization techniques focus on the structure of graphs and do not offer support for dealing with node attributes and edge labels. To enable users to detect relations and patterns in terms of data associated with nodes and edges, we present a technique where this data plays a more central role. Nodes and edges are clustered based on associated data. Via direct manipulation users can interactively inspect and query the graph. Questions that can be answered include, “which edge types are activated by specific node attributes?” and, “how and from where can I reach specific types of nodes?” To validate our approach we contrast it with current practice. We also provide several examples where our method was used to study transition graphs that model real‐world systems.     

Numerical Entropy Production on Shocks and Smooth Transitions

In this work, we propose an algorithm to measure the numerical entropy production of several high order central schemes. The results obtained on a numerical test indicate that this quantity permits to estimate the local error, in regions of smooth flow, and to detect and locate shocks.     

Phase Transitions on Fixed Connected Graphs and Random Graphs in the Presence of Noise

In this paper, we study the phase transition behavior emerging from the interactions among multiple agents in the presence of noise. We propose a simple discrete-time model in which a group of non-mobile agents form either a fixed connected graph or a random graph process, and each agent, taking bipolar value either +1 or -1, updates its value according to its previous value and the noisy measurements of the values of the agents connected to it. We present proofs for the occurrence of the following phase transition behavior: At a noise level higher than some threshold, the system generates symmetric behavior (vapor or melt of magnetization) or disagreement; whereas at a noise level lower than the threshold, the system exhibits spontaneous symmetry breaking (solid or magnetization) or consensus. The threshold is found analytically. The phase transition occurs for any dimension. Finally, we demonstrate the phase transition behavior and all analytic results using simulations. This result may be found useful in the study of the collective behavior of complex systems under communication constraints.     

Unpopping: Solving the Image-Space Blend Problem for Smooth Discrete LOD Transitions

This paper presents a new, simple and practical algorithm to avoid artifacts when switching between discrete levels of detail (LOD) by smoothly blending LOD representations in image space. We analyse the alternatives of conventional alpha‐blending and so‐called late‐switching (the switching of LODs ‘far enough’ from the eye‐point), widely thought to solve the LOD switching discontinuity problem, and conclude that they either do not work in practice, or defeat the concept of LODs. In contrast we show that our algorithm produces visually pleasing blends for static and animated discrete LODs, for discrete LODs with different types of LOD representations (e.g. billboards and meshes) and even to some extent totally different objects with similar spatial extent, with a very small runtime overhead.     

语义增强的大规模多元图简化可视分析方法

网络图可视化可以有效展示网络节点之间的连接关系,广泛应用于诸多领域,如社交网络、知识图谱、生物基因网络等.随着网络数据规模的不断增加,如何简化表达大规模网络图结构已成为图可视化领域中的研究热点.经典的网络图简化可视化方法主要包括图采样、边绑定和图聚类等技术,在减少大量点线交叉造成的视觉紊乱的基础上,提高用户对大规模网络结构的探索和认知效率.然而,上述方法主要侧重于网络图中的拓扑结构,却较少考虑和利用多元图节点的多维属性特征,难以有效提取和表达语义信息,从而无法帮助用户理解大规模多元网络的拓扑结构与多维属性之间的内在关联,为大规模多元图的认知和理解带来困难.因此,本文提出一种语义增强的大规模多元图简化可视分析方法,首先在基于模块度的图聚类算法基础上提取出网络图的层次结构;其次通过多维属性信息熵的计算和比较分析,对网络层次结构进行自适应划分,筛选出具有最优属性聚集特征的社团;进而设计交互便捷的多个关联视图来展示社团之间的拓扑结构、层次关系和属性分布,从不同角度帮助用户分析多维属性在社团形成和网络演化中的作用.大量实验结果表明,本文方法能够有效简化大规模多元图的视觉表达,可以快速分析不同应用领域大规模多元图的关联结构与语义构成,具有较强的实用性.     

Visualization Mosaics for Multivariate Visual Exploration

We present a new model for creating composite visualizations of multidimensional data sets using simple visual representations such as point charts, scatterplots and parallel coordinates as components. Each visual representation is contained in a tile, and the tiles are arranged in a mosaic of views using a space‐filling slice‐and‐dice layout. Tiles can be created, resized, split or merged using a versatile set of interaction techniques, and the visual representation of individual tiles can also be dynamically changed to another representation. Because each tile is self‐contained and independent, it can be implemented in any programming language, on any platform and using any visual representation. We also propose a formalism for expressing visualization mosaics. A Web‐based implementation called MosaicJS supporting multidimensional visual exploration showcases the versatility of the concept and illustrates how it can be used to integrate visualization components provided by different toolkits.     

An Exploration of Embodied Visual Exploration

International Journal of Computer Vision - Embodied computer vision considers perception for robots in novel, unstructured environments. Of particular importance is the embodied visual exploration...     

Building Roadmaps of Minima and Transitions in Visual Models

Becoming trapped in suboptimal local minima is a perennial problem when optimizing visual models, particularly in applications like monocular human body tracking where complicated parametric models are repeatedly fitted to ambiguous image measurements. We show that trapping can be significantly reduced by building roadmaps of nearby minima linked by transition pathways—paths leading over low mountain passes in the cost surface—found by locating the transition state (codimension-1 saddle point) at the top of the pass and then sliding downhill to the next minimum. We present two families of transition-state-finding algorithms based on local optimization. In eigenvector tracking, unconstrained Newton minimization is modified to climb uphill towards a transition state, while in hypersurface sweeping, a moving hypersurface is swept through the space and moving local minima within it are tracked using a constrained Newton method. These widely applicable numerical methods, which appear not to be known in vision and optimization, generalize methods from computational chemistry where finding transition states is critical for predicting reaction parameters. Experiments on the challenging problem of estimating 3D human pose from monocular images show that our algorithms find nearby transition states and minima very efficiently, but also underline the disturbingly large numbers of minima that can exist in this and similar model based vision problems.     

Pathfinder: Visual Analysis of Paths in Graphs

The analysis of paths in graphs is highly relevant in many domains. Typically, path‐related tasks are performed in node‐link layouts. Unfortunately, graph layouts often do not scale to the size of many real world networks. Also, many networks are multivariate, i.e., contain rich attribute sets associated with the nodes and edges. These attributes are often critical in judging paths, but directly visualizing attributes in a graph layout exacerbates the scalability problem. In this paper, we present visual analysis solutions dedicated to path‐related tasks in large and highly multivariate graphs. We show that by focusing on paths, we can address the scalability problem of multivariate graph visualization, equipping analysts with a powerful tool to explore large graphs. We introduce Pathfinder, a technique that provides visual methods to query paths, while considering various constraints. The resulting set of paths is visualized in both a ranked list and as a node‐link diagram. For the paths in the list, we display rich attribute data associated with nodes and edges, and the node‐link diagram provides topological context. The paths can be ranked based on topological properties, such as path length or average node degree, and scores derived from attribute data. Pathfinder is designed to scale to graphs with tens of thousands of nodes and edges by employing strategies such as incremental query results. We demonstrate Pathfinder's fitness for use in scenarios with data from a coauthor network and biological pathways.     

论视觉识别系统中辅助图形的设计

辅助图形的设计是对标准图形在使用时的有益补充和延伸,目地是强化受众者对企业形象多角度、多层次的感受。在传递信息的读图时代,怎样在纷繁复杂中找到最佳的受众渠道,为企业形象添光彩,辅助图形的作用不可估量。     

Smooth Transitions Between Parallel Coordinates and Scatter Plots via Polycurve Star Plots

This paper presents new techniques for seamlessly transitioning between parallel coordinate plots, star plots, and scatter plots. The star plot serves as a mediator visualization between parallel coordinate plots and scatter plots since it uses lines to represent data items as parallel coordinates do and can arrange axes orthogonally as used for scatter plots. The design of the transitions also motivated a new variant of the star plot, the polycurve star plot, that uses curved lines instead of straight ones and has advantages both in terms of space utilization and the detection of clusters. Furthermore, we developed a geometrically motivated method to embed scatter points from a scatter plot into star plots and parallel coordinate plots to track the transition of structural information such as clusters and correlations between the different plot types. The integration of our techniques into an interactive analysis tool for exploring multivariate data demonstrates the advantages and utility of our approach over a multi-view approach for scatter plots and parallel coordinate plots, which we confirmed in a user study and concrete usage scenarios.     

含不可观变迁事件图的状态估计器设计

针对基于事件图建模的离散事件系统中存在的不可观变迁的状态估计问题,本文提出前向、后 向可观路径的概念对其进行了详细分析,并给出一种估计器设计方法,用来根据关联矩阵行变换进行不可观 变迁状态的估计．分析了估计器的观测代价,在此基础上给出了寻找最优估计器的方法,并利用最优估计器 计算不可观变迁的状态估计范围．最后通过实例证明了该方法能够有效地进行不可观变迁的状态估计．     

Cascading Model in Underwater Wireless Sensors using Routing Policy for State Transitions

Underwater wireless sensors communicate information from different depth of water column to sink positioned on the surface of water. Monotonically increasing data traffic occurs as it reaches surface of water column causing an imbalance of routing and resulting in congested nodes. Underwater wireless sensor node can be either in normal, overloaded state, congested state. The purpose of this work is achieving control over normal and isolated states. The congested node use the permutation of links and it observe the transition probabilities for routing. Then it determines the state of node based on the immediate reward and the action to be taken within the time episode. Thus nodes are isolated and recovered after the time episode. Improved estimate in forwarding scenario of overloaded and isolating nodes is observed via simulation results in NS2 simulator.     

Symbolic State Exploration

The exploration of the state space of the model is at the heart of model checking. Symbolic algorithms often use Binary Decision Diagrams for the representation of sets of states, and have to pay attention to the size of the BDDs. We review the techniques that have been proposed for this task in the framework of closed sequences of monotonic functions.     

Multiscale Visualization and Exploration of Large Bipartite Graphs

A bipartite graph is a powerful abstraction for modeling relationships between two collections. Visualizations of bipartite graphs allow users to understand the mutual relationships between the elements in the two collections, e.g., by identifying clusters of similarly connected elements. However, commonly‐used visual representations do not scale for the analysis of large bipartite graphs containing tens of millions of vertices, often resorting to an a‐priori clustering of the sets. To address this issue, we present the Who's‐Active‐On‐What‐Visualization (WAOW‐Vis) that allows for multiscale exploration of a bipartite social‐network without imposing an a‐priori clustering. To this end, we propose to treat a bipartite graph as a high‐dimensional space and we create the WAOW‐Vis adapting the multiscale dimensionality‐reduction technique HSNE. The application of HSNE for bipartite graph requires several modifications that form the contributions of this work. Given the nature of the problem, a set‐based similarity is proposed. For efficient and scalable computations, we use compressed bitmaps to represent sets and we present a novel space partitioning tree to efficiently compute similarities; the Sets Intersection Tree. Finally, we validate WAOW‐Vis on several datasets connecting Twitter‐users and ‐streams in different domains: news, computer science and politics. We show how WAOW‐Vis is particularly effective in identifying hierarchies of communities among social‐media users.     

Visual Analytics for the Exploration of Tumor Tissue Characterization

Tumors are heterogeneous tissues consisting of multiple regions with distinct characteristics. Characterization of these intra‐tumor regions can improve patient diagnosis and enable a better targeted treatment. Ideally, tissue characterization could be performed non‐invasively, using medical imaging data, to derive per voxel a number of features, indicative of tissue properties. However, the high dimensionality and complexity of this imaging‐derived feature space is prohibiting for easy exploration and analysis ‐ especially when clinical researchers require to associate observations from the feature space to other reference data, e.g., features derived from histopathological data. Currently, the exploratory approach used in clinical research consists of juxtaposing these data, visually comparing them and mentally reconstructing their relationships. This is a time consuming and tedious process, from which it is difficult to obtain the required insight. We propose a visual tool for: (1) easy exploration and visual analysis of the feature space of imaging‐derived tissue characteristics and (2) knowledge discovery and hypothesis generation and confirmation, with respect to reference data used in clinical research. We employ, as central view, a 2D embedding of the imaging‐derived features. Multiple linked interactive views provide functionality for the exploration and analysis of the local structure of the feature space, enabling linking to patient anatomy and clinical reference data. We performed an initial evaluation with ten clinical researchers. All participants agreed that, unlike current practice, the proposed visual tool enables them to identify, explore and analyze heterogeneous intra‐tumor regions and particularly, to generate and confirm hypotheses, with respect to clinical reference data.     

Monotonicity Inference for Higher-Order Formulas

Formulas are often monotonic in the sense that satisfiability for a given domain of discourse entails satisfiability for all larger domains. Monotonicity is undecidable in general, but we devised three calculi that infer it in many cases for higher-order logic. The third calculus has been implemented in Isabelle’s model finder Nitpick, where it is used both to prune the search space and to soundly interpret infinite types with finite sets, leading to dramatic speed and precision improvements.