دانلود کتاب هوش مصنوعی عملی با استفاده از پایتون – هوش مصنوعی با استفاده از کتابخانه‌ها، افزونه‌ها و چارچوب‌های مبتنی بر پایتون

Explainable artificial intelligent (XAI) is a current need as more and more AI models are in production to generate business decisions. Thus, many users are also getting impacted by these decisions. One user may get favorably or unfavorably impacted. As a result, it’s important to know the key features leading to these decisions. It is often argued that AI models are quite black-box in nature because the AI model’s decisions cannot be explained, hence the adoptability of AI models is quite slow in the industry. This book is an attempt at unboxing the so-called black-box models to increase the adaptability, interpretability, and explainability of the decisions made by AI algorithms using frameworks such as Python XAI libraries, TensorFlow 2.0+, Keras, and custom frameworks using Python Wrappers. The objective of this book is to explain the AI models in simple language using the above mentioned frameworks. Model interpretability and explainability are the key focuses of this book. There are mathematical formulas and methods that are typically used to explain a decision made by an AI model. You will be provided with software library methods, classes, frameworks, and functions and how to use them for model explainability, transparency, reliability, ethics, bias, and interpretability. If a human being can understand the reasons behind the decision made by the AI model, it will give much more power to the user to make amendments and recommendations. There are two different kinds of users: business users and practitioners. The business user wants the explainability in simple language without any statistical or mathematical terms. The practitioner wants to know the explainability from the computational point of view. This book is an attempt at balancing both needs when generating explanations.