内容简介
人工智能（AI），尤其是生成式语言模型和生成式人工智能（AIGC）模型，正以惊人的速度改变着我们的世界。驾驭这股潮流的关键，莫过于探究自然语言处理（NLP）技术的深奥秘境。本书将带领读者踏上一段扣人心弦的探索之旅，让其亲身感受，并动手搭建语言模型。本书主要内容包括N-Gram，词袋模型（BoW），Word2Vec（W2V），神经概率语言模型（NPLM），循环神经网络（RNN），Seq2Seq（S2S），注意力机制，Transformer，从初代GPT到ChatGPT再到GPT-4等一系列突破性技术的诞生与演进。
本书将以生动活泼的笔触，将枯燥的技术细节化作轻松幽默的故事和缤纷多彩的图画，引领读者穿梭于不同技术的时空，见证自然语言处理技术的传承、演进与蜕变。在这场不断攀登技术新峰的奇妙之旅中，读者不仅能深入理解自然语言处理技术的核心原理，还能自己动手，从零开始搭建起一个又一个语言模型。
无论你是在校学生还是人工智能从业者，这本书都将成为一盏明灯，照亮你探索人工智能无限奥秘的道路。
作者简介
黄佳，笔名咖哥，新加坡科技研究局人工智能研究员。主攻方向为NLP大模型的研发与应用、持续学习、AI in FinTech。黄佳深耕人工智能领域多年，积累了丰富的科研项目和政府、银行、能源、医疗等领域AI项目落地实战经验，目前正与PlatoX.AI展开富有前景的技术合作。曾著有《零基础学机器学习》《数据分析咖哥十话》等多部畅销书。同时，在极客时间开设专栏《零基础实战机器学习》《LangChain实战课》，在深蓝学院开设视频课程“生成式预训练语言模型：理论与实战”。
Content introduction
Artificial intelligence (AI), especially generative language models and Generative Artificial Intelligence (AIGC) models, is changing our world at an alarming rate. The key to riding this trend is to explore the deep mysteries of natural language processing (NLP) technology. This book will take the reader on a fascinating journey of discovery, experience, and build a language model. The main contents of this book include N-Gram, Bag of Words Model (BoW), Word2Vec (W2V), Neural probabilistic language Model (NPLM), recurrent neural Network (RNN), Seq2Seq (S2S), attention mechanism, Transformer, and so on. The birth and evolution of a series of breakthrough technologies from the original GPT to ChatGPT to GPT-4.
This book will use lively brushstrokes to turn dull technical details into humorous stories and colorful pictures, leading readers to shuttle through the time and space of different technologies, and witness the inheritance, evolution and transformation of natural language processing technology. In this amazing journey to new peaks, readers will not only gain a deep understanding of the core principles of natural language processing technology, but also build language models from scratch.
Whether you are a student in school or an AI practitioner, this book will serve as a beacon to illuminate your path to explore the infinite mysteries of AI.
About the author
Huang Jia, who goes by the pen name Kaga, is an artificial intelligence researcher at Singapore's Agency for Science, Technology and Research. His main interests are the development and application of large NLP models, continuous learning, AI in FinTech. Huang Jia has been deeply engaged in the field of artificial intelligence for many years, and has accumulated a wealth of practical experience in scientific research projects and AI projects in government, banking, energy, medical and other fields, and is currently launching promising technical cooperation with PlatoX.AI. He has written a number of bestsellers such as "Machine Learning based on zero" and "Data Analysis Kako Ten Words". At the same time, in geek time, he opened the column "Zero-based Actual Combat Machine Learning" and "LangChain Actual Combat Course", and opened the video course "Generative pre-training Language Model: Theory and Practice" in Deep Blue College.
热门标签
书单推荐