For five years, Konrad has imprisoned himself and his crippled wife in an abandoned lime works where he’s conducted odd auditory experiments and prepared to write his masterwork, The Sense of Hearing. As the story begins, he’s just blown the head off his wife with the Mannlicher carbine she kept strapped to her wheelchair. The murder and the bizarre life that led to it are the subject of a mass of hearsay related by an unnamed life-insurance salesman in a narrative as mazy, byzantine, and mysterious as the lime works—Konrad’s sanctuary and tomb.

第22回木村伊兵衛写真賞受賞作品集の復刊。

But it would be unfair to the reader to reveal what happens when a gang of professional crooks gets wind of the scheme and moves to muscle in on this bettors' dream of a long-odds situation. Worked out with all the meticulous detail, terror, and suspense of a nightmare, the tale is, on one level, comparable to a Graham Greene thriller; on another, it explores a group of people, their relationships fears, and loves. For as Leslie A. Fiedler says in his introduction, "John Hawkes.. . makes terror rather than love the center of his work, knowing all the while, of course, that there can be no terror without the hope for love and love's defeat . . . ."

This book is about making machine learning models and their decisions interpretable. After exploring the concepts of interpretability, you will learn about simple, interpretable models such as decision trees, decision rules and linear regression. Later chapters focus on general model-agnostic methods for interpreting black box models like feature importance and accumulated local effects and explaining individual predictions with Shapley values and LIME. All interpretation methods are explained in depth and discussed critically. How do they work under the hood? What are their strengths and weaknesses? How can their outputs be interpreted? This book will enable you to select and correctly apply the interpretation method that is most suitable for your machine learning project.