دانلود کتاب هوش مصنوعی مولد برای همه -یادگیری عمیق، پردازش زبان طبیعی، کارشناسی ارشد مدیریت بازرگانی برای کاربردهای عملی خلاقانه

Generative AI, a branch of artificial intelligence, has made remarkable strides in recent years. These advancements have enabled a variety of applications, including the creation of lifelike images, videos, text, and music. The many benefits of generative AI can help different industries improve their processes and boost efficiency. This technology has the potential to automate and enhance human and machine tasks, as well as autonomously execute business and IT processes. However, it is important to remain aware of the risks and limitations associated with generative AI. Therefore, understanding its evolution, techniques, architecture, and potential risks is essential. This book is designed for readers of all levels, from beginners to experienced professionals. It covers fundamental concepts, practical applications, advanced topics, and includes hands-on coding examples. Chapter 1: AI Fundamentals – The chapter introduces artificial intelligence (AI) technology, its terminologies, and its significance across industries. It explains the concepts of machine learning and various ML techniques, such as supervised, unsupervised, semi-supervised and reinforced learning, highlighting their benefits and challenges. The chapter also focuses on the concept of design patterns, explaining its importance in AI applications. Chapter 2: GenAI Foundation – The chapter is to provides a foundation for generative AI technology that is deep neural network and its significance. It aims to explain the concepts of deep learning and various deep learning (DL) techniques such as deep neural network elements, gradient descent, backpropagation, hyper-parameters, performance metrics and highlighting their significance and options. The chapter includes coded examples for supervised deep learning regression and classification problems . Chapter 3: GenAI for Images – This chapter provides an overview of computer vision, CNN architectures, and image processing techniques. It covers the basics of image processing, computer vision tasks, real-world applications, and the evolution of deep learning architectures. Additionally, it explores representation learning, including autoencoders and variational autoencoders, for computer vision tasks. The chapter discusses encoderdecoder architectures, encoding and decoding mechanisms, and coded examples such as U-Net for image segmentation and VAE for image generation Chapter 4: Transforming Images with GenAI – The chapter provides an overview of generative modeling, covering different types of explicit and implicit modeling methods. It delves into generative adversarial networks (GANs), discussing their architecture, variations, and applications including example for VQGAN and CLIP. Diffusion processes and diffusion models are explained, including their architecture, training process, and the latent diffusion model for image processing (MNST example). Chapter 5: GenAI for Text – The chapter provides an overview of NLP technology, its key concepts, and its applications in various industries. It explains text pre-processing and post-processing techniques, and discusses different deep learning architectures for NLP tasks, including traditional models like RNN and LSTM, and more recent transformer architectures. The chapter focuses on different types of transformer architectures, their components in details, and their applications in tasks like summarization, question answering, and speech-to-text recognition. The chapter includes coded examples to fine tune GPT model for new tasks and Text summarization task. Chapter 6: ChatGPT – In this chapter, we will look into ChatGPT essentials, its features, integrations and addons along with image generation and speech to text examples. Further, we will understand the importance of prompts, their types, frameworks, and examples, and delve into the prompt engineering process. Finally, we will create a prompt Playground application for enterprises. Chapter 7: Large Language Model Frameworks – The objective of this chapter is to provide a overview of LangChain framework, its components, and examples. We will build three prototypes 1.) use llms to chat with your excel documents 2.) Retrieve contextual information using Retrieval Augmented generation (RAG) from customer knowledge base 3.) create a chat application that can query medical research journals and extract property graph out of it. Chapter 8: Large Language Model Operations – The objective of this chapter is to go through complete LLM system lifecycle, starting from data preparation, pre-training, benchmarking, experimentation, AI alignment, model experimentation, model serving, validation, security and monitoring. This chapter aims to explain the concepts of LLMOps, including various optimization techniques to improve the performance and efficiency of the LLM system as a whole. Chapter 9: Generative AI for Enterprise – The chapter provides an overview of how GenAI technology is leveraged by enterprises for their business improvements. This chapter covers various enterprise use cases, GenAI products and their capabilities. It also focusses on risks related to AI models, GenAI model vulnerabilities and factors that enterprise must consider while implementing GenAI projects. It also gives foundation to responsible AI, Its applicability in enterprise systems and the examples. Chapter 10: Advances and Sustainability in Generative AI – The chapter helps to understand the advancements happening in generative AI in terms of business, technology and regulation. We will also look into the types of business strategy that an enterprise must adopt to leverage GenAI’s potential to their competitive advantage. Further we will look into the environmental impact due to training & inferencing of GenAI models, way to measure and mitigate the same.