相關主題
商品描述
This book serves as an introduction to the science and applications of Large Language Models (LLMs). You'll discover the common thread that drives some of the most revolutionary recent applications of artificial intelligence (AI): from conversational systems like ChatGPT or BARD, to machine translation, summary generation, question answering, and much more.
At the heart of these innovative applications is a powerful and rapidly evolving discipline, natural language processing (NLP). For more than 60 years, research in this science has been focused on enabling machines to efficiently understand and generate human language. The secrets behind these technological advances lie in LLMs, whose power lies in their ability to capture complex patterns and learn contextual representations of language. How do these LLMs work? What are the available models and how are they evaluated? This book will help you answer these and many other questions. With a technical but accessible introduction:
- You will explore the fascinating world of LLMs, from its foundations to its most powerful applications
- You will learn how to build your own simple applications with some of the LLMs
Designed to guide you step by step, with six chapters combining theory and practice, along with exercises in Python on the Colab platform, you will master the secrets of LLMs and their application in NLP.
From deep neural networks and attention mechanisms, to the most relevant LLMs such as BERT, GPT-4, LLaMA, Palm-2 and Falcon, this book guides you through the most important achievements in NLP. Not only will you learn the benchmarks used to evaluate the capabilities of these models, but you will also gain the skill to create your own NLP applications. It will be of great value to professionals, researchers and students within AI, data science and beyond.
商品描述(中文翻譯)
本書作為大型語言模型（LLMs）科學及應用的入門介紹。您將發現驅動一些最具革命性人工智慧（AI）應用的共同主題：從像 ChatGPT 或 BARD 這樣的對話系統，到機器翻譯、摘要生成、問題回答等更多應用。
這些創新應用的核心是一個強大且快速發展的學科——自然語言處理（NLP）。在過去的60多年中，這一科學的研究一直專注於使機器能夠有效理解和生成自然語言。這些技術進步的秘密在於 LLMs，其力量在於捕捉複雜模式和學習語言的上下文表示。這些 LLMs 是如何運作的？有哪些可用的模型，如何評估它們？本書將幫助您回答這些及其他許多問題。提供一個技術但易於理解的介紹：
- 您將探索 LLMs 的迷人世界，從其基礎到最強大的應用
- 您將學習如何使用一些 LLMs 建立自己的簡單應用
本書設計為逐步指導，包含六個結合理論與實踐的章節，以及在 Colab 平台上的 Python 練習，您將掌握 LLMs 的秘密及其在 NLP 中的應用。
從深度神經網絡和注意力機制，到最相關的 LLMs，如 BERT、GPT-4、LLaMA、Palm-2 和 Falcon，本書將引導您了解 NLP 中最重要的成就。您不僅會學習用於評估這些模型能力的基準，還將獲得創建自己 NLP 應用的技能。這對於 AI、數據科學及其他領域的專業人士、研究人員和學生將具有很大的價值。
作者簡介
John Atkinson-Abutridy, received a PhD in Artificial Intelligence from the University of Edinburgh in Scotland, and currently is a full professor at the Faculty of Engineering and Sciences at Universidad Adolfo Ibáñez in Santiago, Chile. Over the years, he has also held full-time academic positions at various Chilean universities and abroad as a visiting professor and researcher at university and research centers in Europe (France, UK), Unites States (MIT, IBM T.J. Watson) and various Latin American universities. Dr. Atkinson-Abutridy's primary research interests span into Natural Language Processing, textual analytics, artificial intelligence, and bio-inspired computing. His academic career includes the publication of nearly one hundred scientific papers and the authorship of two books. Recently, he has been at the forefront of numerous scientific and technological projects at national and international level, serving as an AI consultant for many companies and founding AI-Empowered. In recognition of his substantial contributions to the field of computer science, Dr. Atkinson-Abutridy received the Senior Member Award by the Association for Computing Machinery (ACM) in the United States in 2010. Among his notable accomplishments, he pioneered the first worldwide web-based natural language dialogue model in 2005, a precursor to today's ChatGPT system. In 2023, he released the second edition of his book, "Text Analytics: An Introduction to the Science and Applications of Unstructured Information Analysis" (Taylor & Francis, USA), which was recognized as the top choice in the text mining category by the Book Authority organization.
作者簡介(中文翻譯)
約翰·阿特金森-阿布特里迪（John Atkinson-Abutridy）於蘇格蘭愛丁堡大學獲得人工智慧博士學位，目前是智利聖地亞哥阿道夫·伊巴涅斯大學工程與科學學院的全職教授。多年來，他在智利的多所大學及國外擔任全職學術職位，並作為訪問教授和研究員在歐洲（法國、英國）、美國（麻省理工學院、IBM T.J. Watson）及多所拉丁美洲大學的研究中心工作。阿特金森-阿布特里迪博士的主要研究興趣涵蓋自然語言處理、文本分析、人工智慧及生物啟發計算。他的學術生涯包括發表近百篇科學論文及撰寫兩本書籍。最近，他在國內外的多個科學與技術項目中處於前沿，擔任多家公司的人工智慧顧問，並創立了AI-Empowered。因其對計算機科學領域的重大貢獻，阿特金森-阿布特里迪博士於2010年獲得美國計算機協會（ACM）頒發的資深會員獎。值得一提的是，他於2005年開創了全球首個基於網路的自然語言對話模型，這是今日ChatGPT系統的前身。2023年，他發行了其著作《文本分析：非結構化信息分析的科學與應用導論》（Text Analytics: An Introduction to the Science and Applications of Unstructured Information Analysis，泰勒與法蘭西斯出版社，USA）的第二版，該書在文本挖掘類別中被《書籍權威》（Book Authority）組織評選為最佳選擇。
