Search

A common feature of many approaches to modeling sensory statistics is an emphasis on capturing the'average.'From early representations in the brain, to highly abstracted class categories in machine learning for classification tasks, central-tendency models based on the Gaussian distribution are a seemingly natural and obvious choice for modeling sensory data. However, insights from neuroscience, psychology, and computer vision suggest an alternate strategy: preferentially focusing representational resources on the extremes of the distribution of sensory inputs. The notion of treating extrema near a decision boundary as features is not necessarily new, but a comprehensive statistical theory of recognition based on extrema is only now just emerging in the computer vision literature. This book begins by introducing the statistical Extreme Value Theory (EVT) for visual recognition. In contrast to central-tendency modeling, it is hypothesized that distributions near decision boundaries form a more powerful model for recognition tasks by focusing coding resources on data that are arguably the most diagnostic features. EVT has several important properties: strong statistical grounding, better modeling accuracy near decision boundaries than Gaussian modeling, the ability to model asymmetric decision boundaries, and accurate prediction of the probability of an event beyond our experience. The second part of the book uses the theory to describe a new class of machine learning algorithms for decision making that are a measurable advance beyond the state-of-the-art. This includes methods for post-recognition score analysis, information fusion, multi-attribute spaces, and calibration of supervised machine learning algorithms.

This book introduces quantitative intertextuality, a new approach to the algorithmic study of information reuse in text, sound and images. Employing a variety of tools from machine learning, natural language processing, and computer vision, readers will learn to trace patterns of reuse across diverse sources for scholarly work and practical applications.The respective chapters share highly novel methodological insights in order to guide the reader through the basics of intertextuality. In Part 1, “Theory”, the theoretical aspects of intertextuality are introduced, leading to a discussion of how they can be embodied by quantitative methods. In Part 2, “Practice”, specific quantitative methods are described to establish a set of automated procedures for the practice of quantitative intertextuality. Each chapter in Part 2 begins with a general introduction to a major concept (e.g., lexical matching, sound matching, semantic matching), followed by a casestudy (e.g., detecting allusions to a popular television show in tweets, quantifying sound reuse in Romantic poetry, identifying influences in fan faction by thematic matching), and finally the development of an algorithm that can be used to reveal parallels in the relevant contexts.Because this book is intended as a “gentle” introduction, the emphasis is often on simple yet effective algorithms for a given matching task. A set of exercises is included at the end of each chapter, giving readers the chance to explore more cutting-edge solutions and novel aspects to the material at hand. Additionally, the book's companion website includes software (R and C++ library code) and all of the source data for the examples in the book, as well as supplemental content (slides, high-resolution images, additional results) that may prove helpful for exploring the different facets of quantitative intertextuality that are presented in each chapter.Given its interdisciplinary nature, the book will appeal to a broad audience. From practitioners specializing in forensics to students of cultural studies, readers with diverse backgrounds (e.g., in the social sciences, natural language processing, or computer vision) will find valuable insights.

A common feature of many approaches to modeling sensory statistics is an emphasis on capturing the'average.'From early representations in the brain, to highly abstracted class categories in machine learning for classification tasks, central-tendency models based on the Gaussian distribution are a seemingly natural and obvious choice for modeling sensory data. However, insights from neuroscience, psychology, and computer vision suggest an alternate strategy: preferentially focusing representational resources on the extremes of the distribution of sensory inputs. The notion of treating extrema near a decision boundary as features is not necessarily new, but a comprehensive statistical theory of recognition based on extrema is only now just emerging in the computer vision literature. This book begins by introducing the statistical Extreme Value Theory (EVT) for visual recognition. In contrast to central-tendency modeling, it is hypothesized that distributions near decision boundaries form a more powerful model for recognition tasks by focusing coding resources on data that are arguably the most diagnostic features. EVT has several important properties: strong statistical grounding, better modeling accuracy near decision boundaries than Gaussian modeling, the ability to model asymmetric decision boundaries, and accurate prediction of the probability of an event beyond our experience. The second part of the book uses the theory to describe a new class of machine learning algorithms for decision making that are a measurable advance beyond the state-of-the-art. This includes methods for post-recognition score analysis, information fusion, multi-attribute spaces, and calibration of supervised machine learning algorithms.