Revised the abstract

Co-Authored-By: ELSuitorHarvard <[email protected]>
Co-Authored-By: Elias Nuwara <[email protected]>

_quarto.yml

@@ -28,8 +28,7 @@ book:
   page-navigation: true
   title: "MACHINE LEARNING SYSTEMS"
   subtitle: "with TinyML"
-  abstract: Machine Learning Systems with TinyML offers an introduction to end-to-end machine learning pipelines by focusing on their implementation on resource-constrained devicesds. TinyML provides an accessible lens through which to cover general principles and best practices for applying machine learning. As edge computing grows, performant TinyML enables AI on embedded devices. This book provides the expertise to deploy complete machine learning systems - from data preparation to model training, optimization, acceleration and production deployment - seen through the lens of tinyML applications. The book covers important topics like efficient neural network architectures, hardware-aware training, model compression, benchmarking, and on-device learning. Additional chapters highlight advances like on-device data generation and considerations around reliability, privacy, security, and responsible AI. With concrete use cases and hands-on examples, readers will learn to implement end-to-end machine learning workflows on embedded devices. Overall, by grounding the introduction in tinyML applications, the book enables readers to develop specialized knowledge while also learning general concepts for applying machine learning systems to transform edge devices and IoT.
-
+  abstract: Machine Learning Systems with TinyML" offers an accessible entry into the fascinating world of computing that enables machine learning, with a special focus on its implementation in resource-constrained environments using TinyML. Serving as an ideal starting point for newcomers to the field, this book provides insight into the principles and best practices in machine learning through the practical lens of TinyML. As edge computing continues to expand, the importance of efficient ML applications in such environments grows. This book aims to demystify the complex process of deploying complete machine learning systems, encompassing everything from initial data preparation and model training to optimization, acceleration, and eventual deployment. We explore key topics such as efficient neural network architectures, hardware-aware training, model compression, benchmarking, and on-device learning. Additionally, the text delves into emerging areas like on-device data generation, while addressing crucial considerations of reliability, privacy, security, and responsible AI in resource-constrained environments. By grounding these concepts in TinyML, the book imparts our specialized knowledge and offers an introduction to general machine learning techniques. It strives to ensure that readers are well-equipped to transform edge devices and IoT through the application of machine learning.
 
   search: true
   repo-url: https://github.com/harvard-edge/cs249r_book