Machine Learning: The New AI by Ethem Alpaydin. A Fulbright scholar, Ethem holds a Ph.D. in Computer Engineering from Ecole Polytechnique Fédérale de Lausanne. He has held visiting research positions at University of California, Berkeley, and MIT.
Ethem is delivering an exceptional overview of machine learning. If you want to understand the foundations of machine learning without any programming details, this is the perfect book. The math and statistics are delivered at a conceptual level. Anyone can follow along. He provides a solid foundation addressing algorithms, artificial intelligence, and neural networks. Again for anyone interested, this book is not technical. You will not be overwhelmed, but rather inspired to learn.
Today, Machine Learning (ML) certainly is the most popular subset of artificial intelligence. With ML certainly now a core AI service, we can more easily understand the growing range of ML apps we use everyday.
This includes product recommendations to voice recognition. Spread across just seven chapters, readers will come to understand ML, Statistics and Data Analytics. However chapter four: Neural Networks and Deep Learning is a strong delivery of ML’s core services. This is perhaps the most important chapter for readers new to ML. Ethem provides the much needed context that the foundations were first tested in 1946. This helps set a level playing field in following onto neural networks and the core of deep learning.
Teach machines to think
Many will begin to understand ML as not a service where you write programs, but rather collect data. I am reminded of How Smart Machines Think by Sean Gerrish.
Machine learning will help us make sense of an increasingly complex world. Already we are exposed to more data than what our sensors can cope with or our brains can process.
pg. 146
So, yes big data is getting a lot bigger and we need ML to be able to analyze data collected into large data lakes. Ethem offers impactful stories how we have evolved from crunching numbers on large mainframes to simply finding even more computing power in our mobile phones. Ethem is also addressing very well the use of machine learning algorithms for pattern recognition and reinforcement learning. These books also address this area:
Is there anything missing? I would have enjoyed learning his views regarding Automated Machine Learning AutoML and Machine Learning Operations MLOps.
In conclusion, this book is for understanding the concepts of machine learning and the math and statistics at a good conceptual level for any reader. Ethem has written an excellent book that will provide a great introduction to ML for many readers.