No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Artificial Intelligence in Visual Analytics
Abstract
Visual Analytics that combines automated methods with information visualization has emerged as a powerful approach to analytical reasoning. The integration of artificial intelligence techniques into Visual Analytics has enhanced its capabilities but also presents challenges related to interpretability, explainability, and decision-making processes. Visual Analytics may use artificial intelligence methods to provide enhanced and more powerful analytical reasoning capabilities. Furthermore, Visual Analytics can be used to interpret black-box artificial intelligence models and provide a visual explanation of those models. In this paper, we provide an overview of the state-of-the-art of artificial intelligence techniques used in Visual Analytics, focusing on both explainable artificial intelligence in Visual Analytics and the human knowledge generation process through Visual Analytics. We review explainable artificial intelligence approaches in Visual Analytics and propose a revised Visual Analytics model for Explainable artificial intelligence based on an existing model. We then conduct a screening review of artificial intelligence methods in Visual Analytics from two time periods to highlight recently used artificial intelligence approaches in Visual Analytics. Based on this review, we propose a revised task model for tasks in Visual Analytics. Our contributions include a state-of-the-art review of explainable artificial intelligence in Visual Analytics, a revised model for creating explainable artificial intelligence through Visual Analytics, a screening review of recent artificial intelligence methods in Visual Analytics, and a revised task model for generic tasks in Visual Analytics.
This chapter investigates the transformative impact of technological improvements on traditional travel agencies, a topic that is becoming increasingly important in the fast-changing travel industry. Through the previous related literature, this study aims to shed light on how travel firms are adjusting to the changing demands and tastes of travellers in the digital era by studying the integration of artificial intelligence, virtual and augmented reality, and data-driven personalisation. So, the paper aims to identify travel agencies' challenges and opportunities by analysing digital transformation's impact on old business models. In addition, it investigates the shifting behaviour of travellers, who are increasingly looking for ease and personalised experiences through internet platforms. Finally, this chapter emphasises the significance of travel agencies embracing digital innovation, streamlining operations, and developing immersive travel experiences to maintain their competitive edge in a constantly moving market.
Integrating artificial intelligence (AI) and machine learning (ML) methods with interactive visualization is a research area that has evolved for years. With the rise of AI applications, the combination of AI/ML and interactive visualization is elevated to new levels of sophistication and has become more widespread in many domains. Such application drive has led to a growing trend to bridge the gap between AI/ML and visualizations. This chapter summarizes the current research trend and provides foresight to future research direction in integrating AI/ML and visualization. It investigates different areas of integrating the named disciplines, starting with visualization in ML, visual analytics, visual-enabled machine learning, natural language processing, and multidimensional visualization and AI to illustrate the research trend towards visual knowledge discovery. Each section of this chapter presents the current research state along with problem statements or future directions that allow a deeper investigation of seamless integration of novel AI methods in interactive visualizations.
The growing interest in applying machine and deep learning algorithms in an Outcome-Oriented Predictive Process Monitoring (OOPPM) context has recently fuelled a shift to use models from the explainable artificial intelligence (XAI) paradigm, a field of study focused on creating explainability techniques on top of AI models in order to legitimize the predictions made. Nonetheless, most classification models are evaluated primarily on a performance level, where XAI requires striking a balance between either simple models (e.g. linear regression) or models using complex inference structures (e.g. neural networks) with post-processing to calculate feature importance. In this paper, a comprehensive overview of predictive models with varying intrinsic complexity are measured based on explainability with model-agnostic quantitative evaluation metrics. To this end, explainability is designed as a symbiosis between interpretability and faithfulness and thereby allowing to compare inherently created explanations (e.g. decision tree rules) with post-hoc explainability techniques (e.g. Shapley values) on top of AI models. Moreover, two improved versions of the logistic regression model capable of capturing non-linear interactions and both inherently generating their own explanations are proposed in the OOPPM context. These models are benchmarked with two common state-of-the-art models with post-hoc explanation techniques in the explainability-performance space.
To make predictions and explore large datasets, healthcare is increasingly applying advanced algorithms of artificial intelligence. However, to make well‐considered and trustworthy decisions, healthcare professionals require ways to gain insights in these algorithms' outputs. One approach is visual analytics, which integrates humans in decision‐making through visualizations that facilitate interaction with algorithms. Although many visual analytics systems have been developed for healthcare, a clear overview of their explanation techniques is lacking. Therefore, we review 71 visual analytics systems for healthcare, and analyze how they explain advanced algorithms through visualization, interaction, shepherding, and direct explanation. Based on our analysis, we outline research opportunities and challenges to further guide the exciting rapprochement of visual analytics and healthcare.
This article is categorized under: Application Areas > Health Care
Fundamental Concepts of Data and Knowledge > Explainable AI
Technologies > Visualization
The awareness of emerging trends is essential for strategic decision making because technological trends can affect a firm’s competitiveness and market position. The rise of artificial intelligence methods allows gathering new insights and may support these decision-making processes. However, it is essential to keep the human in the loop of these complex analytical tasks, which, often lack an appropriate interaction design. Including special interactive designs for technology and innovation management is therefore essential for successfully analyzing emerging trends and using this information for strategic decision making. A combination of information visualization, trend mining and interaction design can support human users to explore, detect, and identify such trends. This paper enhances and extends a previously published first approach for integrating, enriching, mining, analyzing, identifying, and visualizing emerging trends for technology and innovation management. We introduce a novel interaction design by investigating the main ideas from technology and innovation management and enable a more appropriate interaction approach for technology foresight and innovation detection.
Deep reinforcement learning (DRL) targets to train an autonomous agent to interact with a pre-defined environment and strives to achieve specific goals through deep neural networks (DNN). Recurrent neural network (RNN) based DRL has demonstrated superior performance, as RNNs can effectively capture the temporal evolution of the environment and respond with proper agent actions. However, apart from the outstanding performance, little is known about how RNNs understand the environment internally and what has been memorized over time. Revealing these details is extremely important for deep learning experts to understand and improve DRLs, which in contrast, is also challenging due to the complicated data transformations inside these models. In this paper, we propose Deep Reinforcement Learning Interactive Visual Explorer (DRLIVE), a visual analytics system to effectively explore, interpret, and diagnose RNN-based DRLs. Having focused on DRL agents trained for different Atari games, DRLIVE accomplishes three tasks: game episode exploration, RNN hidden/cell state examination, and interactive model perturbation. Using the system, one can flexibly explore a DRL agent through interactive visualizations, discover interpretable RNN cells by prioritizing RNN hidden/cell states with a set of metrics, and further diagnose the DRL model by interactively perturbing its inputs. Through concrete studies with multiple deep learning experts, we validated the efficacy of DRLIVE.
The contemporary process-aware information systems possess the capabilities to record the activities generated during the process execution. To leverage these process specific fine-granular data, process mining has recently emerged as a promising research discipline. As an important branch of process mining, predictive business process management, pursues the objective to generate forward-looking, predictive insights to shape business processes. In this study, we propose a conceptual framework sought to establish and promote understanding of decision-making environment, underlying business processes and nature of the user characteristics for developing explainable business process prediction solutions. Consequently, with regard to the theoretical and practical implications of the framework, this study proposes a novel local post-hoc explanation approach for a deep learning classifier that is expected to facilitate the domain experts in justifying the model decisions. In contrary to alternative popular perturbation-based local explanation approaches, this study defines the local regions from the validation dataset by using the intermediate latent space representations learned by the deep neural networks. To validate the applicability of the proposed explanation method, the real-life process log data delivered by the Volvo IT Belgium’s incident management system are used. The adopted deep learning classifier achieves a good performance with the area under the ROC Curve of 0.94. The generated local explanations are also visualized and presented with relevant evaluation measures which are expected to increase the users’ trust in the black-box model.
Artificial Intelligence (AI) has come to prominence as one of the major components of our society, with applications in most aspects of our lives. In this field, complex and highly nonlinear machine learning models such as ensemble models, deep neural networks, and Support Vector Machines have consistently shown remarkable accuracy in solving complex tasks. Although accurate, AI models often are “black boxes” which we are not able to understand. Relying on these models has a multifaceted impact and raises significant concerns about their transparency. Applications in sensitive and critical domains are a strong motivational factor in trying to understand the behavior of black boxes. We propose to address this issue by providing an interpretable layer on top of black box models by aggregating “local” explanations. We present GLocalX, a “local-first” model agnostic explanation method. Starting from local explanations expressed in form of local decision rules, GLocalX iteratively generalizes them into global explanations by hierarchically aggregating them. Our goal is to learn accurate yet simple interpretable models to emulate the given black box, and, if possible, replace it entirely. We validate GLocalX in a set of experiments in standard and constrained settings with limited or no access to either data or local explanations. Experiments show that GLocalX is able to accurately emulate several models with simple and small models, reaching state-of-the-art performance against natively global solutions. Our findings show how it is often possible to achieve a high level of both accuracy and comprehensibility of classification models, even in complex domains with high-dimensional data, without necessarily trading one property for the other. This is a key requirement for a trustworthy AI, necessary for adoption in high-stakes decision making applications.
The explainable artificial intelligence (xAI) is one of the interesting issues that has emerged recently. Many researchers are trying to deal with the subject with different dimensions and interesting results that have come out. However, we are still at the beginning of the way to understand these types of models. The forthcoming years are expected to be years in which the openness of deep learning models is discussed. In classical artificial intelligence approaches, we frequently encounter deep learning methods available today. These deep learning methods can yield highly effective results according to the data set size, data set quality, the methods used in feature extraction, the hyper parameter set used in deep learning models, the activation functions, and the optimization algorithms. However, there are important shortcomings that current deep learning models are currently inadequate. These artificial neural network-based models are black box models that generalize the data transmitted to it and learn from the data. Therefore, the relational link between input and output is not observable. This is an important open point in artificial neural networks and deep learning models. For these reasons, it is necessary to make serious efforts on the explainability and interpretability of black box models.
As a consequence of technological progress, nowadays, one is used to the availability of big data generated in nearly all fields of science. However, the analysis of such data possesses vast challenges. One of these challenges relates to the explainability of methods from artificial intelligence (AI) or machine learning. Currently, many of such methods are nontransparent with respect to their working mechanism and for this reason are called black box models, most notably deep learning methods. However, it has been realized that this constitutes severe problems for a number of fields including the health sciences and criminal justice and arguments have been brought forward in favor of an explainable AI (XAI). In this paper, we do not assume the usual perspective presenting XAI as it should be, but rather provide a discussion what XAI can be . The difference is that we do not present wishful thinking but reality grounded properties in relation to a scientific theory beyond physics.
This article is categorized under:
• Fundamental Concepts of Data and Knowledge > Explainable AI
• Algorithmic Development > Statistics
• Technologies > Machine Learning
Machine learning (ML) models are nowadays used in complex applications in various domains, such as medicine, bioinformatics, and other sciences. Due to their black box nature, however, it may sometimes be hard to understand and trust the results they provide. This has increased the demand for reliable visualization tools related to enhancing trust in ML models, which has become a prominent topic of research in the visualization community over the past decades. To provide an overview and present the frontiers of current research on the topic, we present a State‐of‐the‐Art Report (STAR) on enhancing trust in ML models with the use of interactive visualization. We define and describe the background of the topic, introduce a categorization for visualization techniques that aim to accomplish this goal, and discuss insights and opportunities for future research directions. Among our contributions is a categorization of trust against different facets of interactive ML, expanded and improved from previous research. Our results are investigated from different analytical perspectives: (a) providing a statistical overview, (b) summarizing key findings, (c) performing topic analyses, and (d) exploring the data sets used in the individual papers, all with the support of an interactive web‐based survey browser. We intend this survey to be beneficial for visualization researchers whose interests involve making ML models more trustworthy, as well as researchers and practitioners from other disciplines in their search for effective visualization techniques suitable for solving their tasks with confidence and conveying meaning to their data.
We propose a technique for producing ‘visual explanations’ for decisions from a large class of Convolutional Neural Network (CNN)-based models, making them more transparent and explainable. Our approach—Gradient-weighted Class Activation Mapping (Grad-CAM), uses the gradients of any target concept (say ‘dog’ in a classification network or a sequence of words in captioning network) flowing into the final convolutional layer to produce a coarse localization map highlighting the important regions in the image for predicting the concept. Unlike previous approaches, Grad-CAM is applicable to a wide variety of CNN model-families: (1) CNNs with fully-connected layers (e.g.VGG), (2) CNNs used for structured outputs (e.g.captioning), (3) CNNs used in tasks with multi-modal inputs (e.g.visual question answering) or reinforcement learning, all without architectural changes or re-training. We combine Grad-CAM with existing fine-grained visualizations to create a high-resolution class-discriminative visualization, Guided Grad-CAM, and apply it to image classification, image captioning, and visual question answering (VQA) models, including ResNet-based architectures. In the context of image classification models, our visualizations (a) lend insights into failure modes of these models (showing that seemingly unreasonable predictions have reasonable explanations), (b) outperform previous methods on the ILSVRC-15 weakly-supervised localization task, (c) are robust to adversarial perturbations, (d) are more faithful to the underlying model, and (e) help achieve model generalization by identifying dataset bias. For image captioning and VQA, our visualizations show that even non-attention based models learn to localize discriminative regions of input image. We devise a way to identify important neurons through Grad-CAM and combine it with neuron names (Bau et al. in Computer vision and pattern recognition, 2017) to provide textual explanations for model decisions. Finally, we design and conduct human studies to measure if Grad-CAM explanations help users establish appropriate trust in predictions from deep networks and show that Grad-CAM helps untrained users successfully discern a ‘stronger’ deep network from a ‘weaker’ one even when both make identical predictions. Our code is available at https://github.com/ramprs/grad-cam/, along with a demo on CloudCV (Agrawal et al., in: Mobile cloud visual media computing, pp 265–290. Springer, 2015) (http://gradcam.cloudcv.org) and a video at http://youtu.be/COjUB9Izk6E.
Deep learning has made significant contribution to the recent progress in artificial intelligence. In comparison to traditional machine learning methods such as decision trees and support vector machines, deep learning methods have achieved substantial improvement in various prediction tasks. However, deep neural networks (DNNs) are comparably weak in explaining their inference processes and final results, and they are typically treated as a black-box by both developers and users. Some people even consider DNNs (deep neural networks) in the current stage rather as alchemy, than as real science. In many real-world applications such as business decision, process optimization, medical diagnosis and investment recommendation, explainability and transparency of our AI systems become particularly essential for their users, for the people who are affected by AI decisions, and furthermore, for the researchers and developers who create the AI solutions. In recent years, the explainability and explainable AI have received increasing attention by both research community and industry. This paper first introduces the history of Explainable AI, starting from expert systems and traditional machine learning approaches to the latest progress in the context of modern deep learning, and then describes the major research areas and the state-of-art approaches in recent years. The paper ends with a discussion on the challenges and future directions.
Clustering, the process of grouping together similar items into distinct partitions, is a common type of unsupervised machine learning that can be useful for summarizing and aggregating complex multi-dimensional data. However, data can be clustered in many ways, and there exist a large body of algorithms designed to reveal different patterns. While having access to a wide variety of algorithms is helpful, in practice, it is quite difficult for data scientists to choose and parameterize algorithms to get the clustering results relevant for their dataset and analytical tasks. To alleviate this problem, we built Clustervision, a visual analytics tool that helps ensure data scientists find the right clustering among the large amount of techniques and parameters available. Our system clusters data using a variety of clustering techniques and parameters and then ranks clustering results utilizing five quality metrics. In addition, users can guide the system to produce more relevant results by providing task-relevant constraints on the data. Our visual user interface allows users to find high quality clustering results, explore the clusters using several coordinated visualization techniques, and select the cluster result that best suits their task. We demonstrate this novel approach using a case study with a team of researchers in the medical domain and showcase that our system empowers users to choose an effective representation of their complex data.
Understanding why a model makes a certain prediction can be as crucial as the prediction's accuracy in many applications. However, the highest accuracy for large modern datasets is often achieved by complex models that even experts struggle to interpret, such as ensemble or deep learning models, creating a tension between accuracy and interpretability. In response, various methods have recently been proposed to help users interpret the predictions of complex models, but it is often unclear how these methods are related and when one method is preferable over another. To address this problem, we present a unified framework for interpreting predictions, SHAP (SHapley Additive exPlanations). SHAP assigns each feature an importance value for a particular prediction. Its novel components include: (1) the identification of a new class of additive feature importance measures, and (2) theoretical results showing there is a unique solution in this class with a set of desirable properties. The new class unifies six existing methods, notable because several recent methods in the class lack the proposed desirable properties. Based on insights from this unification, we present new methods that show improved computational performance and/or better consistency with human intuition than previous approaches.
Human-in-the-loop data analysis applications necessitate greater transparency in machine learning models for experts to understand and trust their decisions. To this end, we propose a visual analytics workflow to help data scientists and domain experts explore, diagnose, and understand the decisions made by a binary classifier. The approach leverages "instance-level explanations", measures of local feature relevance that explain single instances, and uses them to build a set of visual representations that guide the users in their investigation. The workflow is based on three main visual representations and steps: one based on aggregate statistics to see how data distributes across correct / incorrect decisions; one based on explanations to understand which features are used to make these decisions; and one based on raw data, to derive insights on potential root causes for the observed patterns. The work is validated through a long-term collaboration with a group of machine learning and healthcare professionals who used our method to make sense of machine learning models they developed. The case study from this collaboration demonstrates that the proposed method helps experts derive useful knowledge about the model and the phenomena it describes and also generate useful hypotheses on how a model can be improved.
Supervised machine learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world? We want models to be not only good, but inter-pretable. And yet the task of interpretation appears underspecified. Papers provide diverse and sometimes non-overlapping motivations for in-terpretability, and offer myriad notions of what attributes render models interpretable. Despite this ambiguity, many papers proclaim inter-pretability axiomatically, absent further explanation. In this paper, we seek to refine the discourse on interpretability. First, we examine the motivations underlying interest in interpretabil-ity, finding them to be diverse and occasionally discordant. Then, we address model properties and techniques thought to confer interpretability, identifying transparency to humans and post-hoc explanations as competing notions. Throughout, we discuss the feasibility and desirability of different notions, and question the oft-made assertions that linear models are interpretable and that deep neural networks are not.
Understanding and interpreting classification decisions of automated image classification systems is of high value in many applications as it allows to verify the reasoning of the system and provides additional information to the human expert. Although machine learning methods are solving very successfully a plethora of tasks, they have in most cases the disadvantage of acting as a black box, not providing any information about what made them arrive at a particular decision. This work proposes a general solution to the problem of understanding classification decisions by pixel-wise decomposition of non- linear classifiers. We introduce a methodology that allows to visualize the contributions of single pixels to predictions for kernel-based classifiers over Bag of Words features and for multilayered neural networks. These pixel contributions can be visualized as heatmaps and are provided to a human expert who can intuitively not only verify the validity of the classification decision, but also focus further analysis on regions of potential interest. We evaluate our method for classifiers trained on PASCAL VOC 2009 images, synthetic image data containing geometric shapes, the MNIST handwritten digits data set and for the pre-trained ImageNet model available as part of the Caffe open source package.
We introduce a novel model-agnostic system that explains the behavior of complex models with high-precision rules called anchors, representing local, "sufficient" conditions for predictions. We propose an algorithm to efficiently compute these explanations for any black-box model with high-probability guarantees. We demonstrate the flexibility of anchors by explaining a myriad of different models for different domains and tasks. In a user study, we show that anchors enable users to predict how a model would behave on unseen instances with less effort and higher precision, as compared to existing linear explanations or no explanations.
The rapid development of Convolutional Neural Networks (CNNs) in recent years has triggered significant breakthroughs in many machine learning (ML) applications. The ability to understand and compare various CNN models available is thus essential. The conventional approach with visualizing each model's quantitative features, such as classification accuracy and computational complexity, is not sufficient for a deeper understanding and comparison of the behaviors of different models. Moreover, most of the existing tools for assessing CNN behaviors only support comparison between two models and lack the flexibility of customizing the analysis tasks according to user needs. This paper presents a visual analytics system, VAC-CNN (
V
isual
A
nalytics for
C
omparing
CNNs
), that supports the in-depth inspection of a single CNN model as well as comparative studies of two or more models. The ability to compare a larger number of (e.g., tens of) models especially distinguishes our system from previous ones. With a carefully designed model visualization and explaining support, VAC-CNN facilitates a highly interactive workflow that promptly presents both quantitative and qualitative information at each analysis stage. We demonstrate VAC-CNN's effectiveness for assisting novice ML practitioners in evaluating and comparing multiple CNN models through two use cases and one preliminary evaluation study using the image classification tasks on the ImageNet dataset.
Semantic segmentation is a critical component in autonomous driving and has to be thoroughly evaluated due to safety concerns. Deep neural network (DNN) based semantic segmentation models are widely used in autonomous driving. However, it is challenging to evaluate DNN-based models due to their black-box-like nature, and it is even more difficult to assess model performance for crucial objects, such as lost cargos and pedestrians, in autonomous driving applications. In this work, we propose
VASS
, a
V
isual
A
nalytics approach to diagnosing and improving the accuracy and robustness of
S
emantic
S
egmentation models, especially for critical objects moving in various driving scenes. The key component of our approach is a context-aware spatial representation learning that extracts important spatial information of objects, such as position, size, and aspect ratio, with respect to given scene contexts. Based on this spatial representation, we first use it to create visual summarization to analyze models' performance. We then use it to guide the generation of adversarial examples to evaluate models' spatial robustness and obtain actionable insights. We demonstrate the effectiveness of
VASS
via two case studies of lost cargo detection and pedestrian detection in autonomous driving. For both cases, we show quantitative evaluation on the improvement of models' performance with actionable insights obtained from
VASS
.
Deep learning (DL) models have achieved impressive performance in various domains such as medicine, finance, and autonomous vehicle systems with advances in computing power and technologies. However, due to the black-box structure of DL models, the decisions of these learning models often need to be explained to end-users. Explainable Artificial Intelligence (XAI) provides explanations of black-box models to reveal the behavior and underlying decision-making mechanisms of the models through tools, techniques, and algorithms. Visualization techniques help to present model and prediction explanations in a more understandable, explainable, and interpretable way. This survey paper aims to review current trends and challenges of visual analytics in interpreting DL models by adopting XAI methods and present future research directions in this area. We reviewed literature based on two different aspects, model usage and visual approaches. We addressed several research questions based on our findings and then discussed missing points, research gaps, and potential future research directions. This survey provides guidelines to develop a better interpretation of neural networks through XAI methods in the field of visual analytics.
We often desire our models to be interpretable as well as accurate. Prior work on optimizing models for interpretability has relied on easy-to-quantify proxies for interpretability, such as sparsity or the number of operations required. In this work, we optimize for interpretability by directly including humans in the optimization loop. We develop an algorithm that minimizes the number of user studies to find models that are both predictive and interpretable and demonstrate our approach on several data sets. Our human subjects results show trends towards different proxy notions of interpretability on different datasets, which suggests that different proxies are preferred on different tasks.
Recently, a significant amount of research has been investigated on interpretation of deep neural networks (DNNs) which are normally processed as black box models. Among the methods that have been developed, local interpretation methods stand out which have the features of clear expression in interpretation and low computation complexity. Different from existing surveys which cover a broad range of methods on interpretation of DNNs, this survey focuses on local interpretation methods with an in-depth analysis of the representative works including the newly proposed approaches. From the perspective of principles, we first divide local interpretation methods into two main categories: model-driven methods and data-driven methods. Then we make a fine-grained distinction between the two types of these methods, and highlight the latest ideas and principles. We further demonstrate the effects of a number of interpretation methods by reproducing the results through open source software plugins. Finally, we point out research directions in this rapidly evolving field.
This textbook presents the main principles of visual analytics and describes techniques and approaches that have proven their utility and can be readily reproduced. Special emphasis is placed on various instructive examples of analyses, in which the need for and the use of visualisations are explained in detail.
The book begins by introducing the main ideas and concepts of visual analytics and explaining why it should be considered an essential part of data science methodology and practices. It then describes the general principles underlying the visual analytics approaches, including those on appropriate visual representation, the use of interactive techniques, and classes of computational methods. It continues with discussing how to use visualisations for getting aware of data properties that need to be taken into account and for detecting possible data quality issues that may impair the analysis. The second part of the book describes visual analytics methods and workflows, organised by various data types including multidimensional data, data with spatial and temporal components, data describing binary relationships, texts, images and video. For each data type, the specific properties and issues are explained, the relevant analysis tasks are discussed, and appropriate methods and procedures are introduced. The focus here is not on the micro-level details of how the methods work, but on how the methods can be used and how they can be applied to data. The limitations of the methods are also discussed and possible pitfalls are identified.
The textbook is intended for students in data science and, more generally, anyone doing or planning to do practical data analysis. It includes numerous examples demonstrating how visual analytics techniques are used and how they can help analysts to understand the properties of data, gain insights into the subject reflected in the data, and build good models that can be trusted. Based on several years of teaching related courses at the City, University of London, the University of Bonn and TU Munich, as well as industry training at the Fraunhofer Institute IAIS and numerous summer schools, the main content is complemented by sample datasets and detailed, illustrated descriptions of exercises to practice applying visual analytics methods and workflows.
Because of its ability to find complex patterns in high dimensional and heterogeneous data, machine learning (ML) has emerged as a critical tool for making sense of the growing amount of genetic and genomic data available. While the complexity of ML models is what makes them powerful, it also makes them difficult to interpret. Fortunately, efforts to develop approaches that make the inner workings of ML models understandable to humans have improved our ability to make novel biological insights. Here, we discuss the importance of interpretable ML, different strategies for interpreting ML models, and examples of how these strategies have been applied. Finally, we identify challenges and promising future directions for interpretable ML in genetics and genomics.
The rise of sophisticated machine learning models has brought accurate but obscure decision systems, which hide their logic, thus undermining transparency, trust, and the adoption of AI in socially sensitive and safety-critical contexts. We introduce a local rule-based explanation method providing faithful explanations of the decision made by a black-box classifier on a specific instance. The proposed method first learns an interpretable, local classifier on a synthetic neighborhood of the instance under investigation, generated by a genetic algorithm. Then it derives from the interpretable classifier an explanation consisting of a decision rule, explaining the factual reasons of the decision, and a set of counterfactuals, suggesting the changes in the instance features that would lead to a different outcome. Experimental results show that the proposed method outperforms existing approaches in terms of the quality of the explanations and of the accuracy in mimicking the black-box.
The awareness of emerging technologies is essential for strategic decision making in enterprises. Emerging and decreasing technological trends could lead to strengthening the competitiveness and market positioning. The exploration, detection and identification of such trends can be essentially supported through information visualization, trend mining and in particular through the combination of those. Commonly, trends appear first in science and scientific documents. However, those documents do not provide sufficient information for analyzing and identifying emerging trends. It is necessary to enrich data, extract information from the integrated data, measure the gradient of trends over time and provide effective interactive visualizations. We introduce in this paper an approach for integrating, enriching, mining, analyzing, identifying and visualizing emerging trends from scientific documents. Our approach enhances the state of the art in visual trend analytics by investigating the entire analysis process and providing an approach for enabling human to explore undetected potentially emerging trends.
(Best Paper Award at IV2019)
The collection of large, complex datasets has become common across a wide variety of domains. Visual analytics tools increasingly play a key role in exploring and answering complex questions about these large datasets. However, many visualizations are not designed to concurrently visualize the large number of dimensions present in complex datasets (e.g. tens of thousands of distinct codes in an electronic health record system). This fact, combined with the ability of many visual analytics systems to enable rapid, ad-hoc specification of groups, or cohorts, of individuals based on a small subset of visualized dimensions, leads to the possibility of introducing selection bias-when the user creates a cohort based on a specified set of dimensions, differences across many other unseen dimensions may also be introduced. These unintended side effects may result in the cohort no longer being representative of the larger population intended to be studied, which can negatively affect the validity of subsequent analyses. We present techniques for selection bias tracking and visualization that can be incorporated into high-dimensional exploratory visual analytics systems, with a focus on medical data with existing data hierarchies. These techniques include: (1) tree-based cohort provenance and visualization, including a user-specified baseline cohort that all other cohorts are compared against, and visual encoding of cohort "drift", which indicates where selection bias may have occurred, and (2) a set of visualizations, including a novel icicle-plot based visualization, to compare in detail the per-dimension differences between the baseline and a user-specified focus cohort. These techniques are integrated into a medical temporal event sequence visual analytics tool. We present example use cases and report findings from domain expert user interviews.
Although deep neural networks have achieved state-of-the-art performance in several artificial intelligence applications in the past decade, they are still hard to understand. In particular, the features learned by deep networks when determining whether a given input belongs to a specific class are only implicitly described concerning a considerable number of internal model parameters. This makes it harder to construct interpretable hypotheses of what the network is learning and how it is learning—both of which are essential when designing and improving a deep model to tackle a particular learning task. This challenge can be addressed by the use of visualization tools that allow machine learning experts to explore which components of a network are learning useful features for a pattern recognition task, and also to identify characteristics of the network that can be changed to improve its performance. We present a review of modern approaches aiming to use visual analytics and information visualization techniques to understand, interpret, and fine-tune deep learning models. For this, we propose a taxonomy of such approaches based on whether they provide tools for visualizing a network's architecture, to facilitate the interpretation and analysis of the training process, or to allow for feature understanding. Next, we detail how these approaches tackle the tasks above for three common deep architectures: deep feedforward networks, convolutional neural networks, and recurrent neural networks. Additionally, we discuss the challenges faced by each network architecture and outline promising topics for future research in visualization techniques for deep learning models.
Clustering is a core building block for data analysis, aiming to extract otherwise hidden structures and relations from raw datasets, such as particular groups that can be effectively related, compared, and interpreted. A plethora of visual-interactive cluster analysis techniques has been proposed to date, however, arriving at useful clusterings often requires several rounds of user interactions to fine-tune the data preprocessing and algorithms. We present a multi-stage Visual Analytics (VA) approach for iterative cluster refinement together with an implementation (SOMFlow) that uses Self-Organizing Maps (SOM) to analyze time series data. It supports exploration by offering the analyst a visual platform to analyze intermediate results, adapt the underlying computations, iteratively partition the data, and to reflect previous analytical activities. The history of previous decisions is explicitly visualized within a flow graph, allowing to compare earlier cluster refinements and to explore relations. We further leverage quality and interestingness measures to guide the analyst in the discovery of useful patterns, relations, and data partitions. We conducted two pair analytics experiments together with a subject matter expert in speech intonation research to demonstrate that the approach is effective for interactive data analysis, supporting enhanced understanding of clustering results as well as the interactive process itself.
The recent successes of deep learning have led to a wave of interest from non-experts. Gaining an understanding of this technology, however, is difficult. While the theory is important, it is also helpful for novices to develop an intuitive feel for the effect of different hyperparameters and structural variations. We describe TensorFlow Playground, an interactive, open sourced visualization that allows users to experiment via direct manipulation rather than coding, enabling them to quickly build an intuition about neural nets.
Despite widespread adoption, machine learning models remain mostly black boxes. Understanding the reasons behind predictions is, however, quite important in assessing trust, which is fundamental if one plans to take action based on a prediction, or when choosing whether to deploy a new model. Such understanding also provides insights into the model, which can be used to transform an untrustworthy model or prediction into a trustworthy one.
In this work, we propose LIME, a novel explanation technique that explains the predictions of any classifier in an interpretable and faithful manner, by learning an interpretable model locally varound the prediction. We also propose a method to explain models by presenting representative individual predictions and their explanations in a non-redundant way, framing the task as a submodular optimization problem. We demonstrate the flexibility of these methods by explaining different models for text (e.g. random forests) and image classification (e.g. neural networks). We show the utility of explanations via novel experiments, both simulated and with human subjects, on various scenarios that require trust: deciding if one should trust a prediction, choosing between models, improving an untrustworthy classifier, and identifying why a classifier should not be trusted.
For an autonomous system to be helpful to humans and to pose no unwarranted risks, it needs to align its values with those of the humans in its environment in such a way that its actions contribute to the maximization of value for the humans. We propose a formal definition of the value alignment problem as {\em cooperative inverse reinforcement learning} (CIRL). A CIRL problem is a cooperative, partial-information game with two agents, human and robot; both are rewarded according to the human's reward function, but the robot does not initially know what this is. In contrast to classical IRL, where the human is assumed to act optimally in isolation, optimal CIRL solutions produce behaviors such as active teaching, active learning, and communicative actions that are more effective in achieving value alignment. We show that computing optimal joint policies in CIRL games can be reduced to solving a POMDP, prove that optimality in isolation is suboptimal in CIRL, and derive an approximate CIRL algorithm.
The early awareness of new technologies and upcoming trends is essential for making strategic decisions in enterprises and research. Trends may signal that technologies or related topics might be of great interest in the future or obsolete for future directions. The identification of such trends premises analytical skills that can be supported through trend mining and visual analytics. Thus the earliest trends or signals commonly appear in science, the investigation of digital libraries in this context is inevitable. However, digital libraries do not provide sufficient information for analyzing trends. It is necessary to integrate data, extract information from the integrated data and provide effective interactive visual analysis tools. We introduce in this paper a model that investigates all stages from data integration to interactive visualization for identifying trends and analyzing the market situation through our visual trend analysis environment. Our approach improves the visual analysis of trends by investigating the entire transformation steps from raw and structured data to visual representations.
We aim to produce predictive models that are not only accurate, but are also
interpretable to human experts. Our models are decision lists, which consist of
a series of if...then... statements (e.g., if high blood pressure, then stroke)
that discretize a high-dimensional, multivariate feature space into a series of
simple, readily interpretable decision statements. We introduce a generative
model called Bayesian Rule Lists that yields a posterior distribution over
possible decision lists. It employs a novel prior structure to encourage
sparsity. Our experiments show that Bayesian Rule Lists has predictive accuracy
on par with the current top algorithms for prediction in machine learning. Our
method is motivated by recent developments in personalized medicine, and can be
used to produce highly accurate and interpretable medical scoring systems. We
demonstrate this by producing an alternative to the CHADS score, actively
used in clinical practice for estimating the risk of stroke in patients that
have atrial fibrillation. Our model is as interpretable as CHADS, but more
accurate.
Problem-solving procedures have been typically aimed at achieving well-defined goals or satisfying straightforward preferences. However, learners and solvers may often generate rich multiattribute results with procedures guided by sets of controls that define different dimensions of quality. We explore methods that enable people to explore and express preferences about the operation of classification models in supervised multiclass learning. We leverage a leave-one-out confusion matrix that provides users with views and real-time controls of a model space. The approach allows people to consider in an interactive manner the global implications of local changes in decision boundaries. We focus on kernel classifiers and show the effectiveness of the methodology on a variety of tasks.
We describe latent Dirichlet allocation (LDA), a generative probabilistic model for collections of discrete data such as text corpora. LDA is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is, in turn, modeled as an infinite mixture over an underlying set of topic probabilities. In the context of text modeling, the topic probabilities provide an explicit representation of a document. We present efficient approximate inference techniques based on variational methods and an EM algorithm for empirical Bayes parameter estimation. We report results in document modeling, text classification, and collaborative filtering, comparing to a mixture of unigrams model and the probabilistic LSI model.
Ripley brings together two crucial ideas in pattern recognition: statistical methods and machine learning via neural networks. He brings unifying principles to the fore, and reviews the state of the subject. Ripley also includes many examples to illustrate real problems in pattern recognition and how to overcome them.
Machine learning is an increasingly used computational tool within human-computer interaction research. While most researchers currently utilize an iterative approach to refining classifier models and performance, we propose that ensemble classification techniques may be a viable and even preferable alternative. In ensemble learning, algo- rithms combine multiple classifiers to build one that is su- perior to its components. In this paper, we present Ensem- bleMatrix, an interactive visualization system that presents a graphical view of confusion matrices to help users under- stand relative merits of various classifiers. EnsembleMatrix allows users to directly interact with the visualizations in order to explore and build combination models. We eva- luate the efficacy of the system and the approach in a user study. Results show that users are able to quickly combine multiple classifiers operating on multiple feature sets to produce an ensemble classifier with accuracy that ap- proaches best-reported performance classifying images in the CalTech-101 dataset.
Visual-interactive cluster analysis provides valuable tools for ef-fectively analyzing large and complex data sets. Due to desirable properties and an inherent predisposition for visualization, the Ko-honen Feature Map (or Self-Organizing Map, or SOM) algorithm is among the most popular and widely used visual clustering tech-niques. However, the unsupervised nature of the algorithm may be disadvantageous in certain applications. Depending on initial-ization and data characteristics, cluster maps (cluster layouts) may emerge that do not comply with user preferences, expectations, or the application context. Considering SOM-based analysis of trajectory data, we propose a comprehensive visual-interactive monitoring and control frame-work extending the basic SOM algorithm. The framework imple-ments the general Visual Analytics idea to effectively combine au-tomatic data analysis with human expert supervision. It provides simple, yet effective facilities for visually monitoring and interac-tively controlling the trajectory clustering process at arbitrary levels of detail. The approach allows the user to leverage existing do-main knowledge and user preferences, arriving at improved cluster maps. We apply the framework on a trajectory clustering prob-lem, demonstrating its potential in combining both unsupervised (machine) and supervised (human expert) processing, in producing appropriate cluster results.
We describe latent Dirichlet allocation (LDA), a generative probabilistic model for collections of discrete data such as text corpora. LDA is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is, in turn, modeled as an infinite mixture over an underlying set of topic probabilities. In the context of text modeling, the topic probabilities provide an explicit representation of a document. We present efficient approximate inference techniques based on variational methods and an EM algorithm for empirical Bayes parameter estimation. We report results in document modeling, text classification, and collaborative filtering, comparing to a mixture of unigrams model and the probabilistic LSI model.
Visualization techniques are of increasing importance in exploring and analyzing large amounts of multidimensional information. One important class of visualization techniques which is particularly interesting for visualizing very large multidimensional data sets is the class of the pixel-oriented techniques. The basic idea of pixel-oriented visualization techniques is to represent as many data objects as possible on the screen at the same time by mapping each data value to a pixel of the screen and arranging the pixels adequately. A number of different pixel-oriented visualization techniques have been proposed in recent years and it has been shown that the techniques are useful for visual data exploration in a number of different application contexts. In this paper, we discuss a number of issues which are of high importance in developing pixel-oriented visualization techniques. The major goal of this article is to provide a formal basis of pixel-oriented visualization techniques and show that the design decisions in developing them can be seen as solutions of well-defined optimization problems. This is true for the mapping of the data values to colors, the arrangement of pixels inside the subwindows, the shape of the subwindows, and the ordering of the dimension subwindows. The paper also discusses the design issues of special variants of pixel-oriented techniques for visualizing large spatial data sets. The optimization functions for the mentioned design decisions are important for the effectiveness of the resulting visualizations. We show this by evaluating the optimization functions and comparing it the results to the visualization obtained in a number of different application.
Opportunities and challenges in explainable artificial intelligence (xai): A survey
- das
A. Das and P. Rad, "Opportunities and challenges in explainable artificial
intelligence (xai): A survey," arXiv, 2020.