TLDR: In this article we will explore machine learning definitions from leading experts and books, so sit back, relax, and enjoy seeing how the field’s brightest minds explain this revolutionary technology!
Ever wondered how different experts describe the magic behind Netflix recommendations or spam filters? Well, it turns out that machine learning definitions vary across books and researchers, each highlighting different aspects of this fascinating field.
In this article, we’ll dive into how the biggest names in AI define machine learning and what their unique perspectives tell us about this transformative technology.
Don’t get too excited – we’re not going to create a superintelligent AI together, but by the end of this article, you’ll understand the key machine learning definitions from multiple perspectives, giving you a richer understanding of what makes computers learn. Let’s jump right in!
The Beginning — Machine Learning Definitions Through Time
Machine learning has evolved dramatically over the decades, and so have the machine learning definitions used to describe it. Let’s explore how different experts have characterized this field.
Arthur Samuel (1959)
One of the earliest and most enduring machine learning definitions comes from Arthur Samuel, who defined it as:
“The field of study that gives computers the ability to learn without being explicitly programmed.”
This remarkably forward-thinking definition emerged before modern computing power made today’s machine learning applications possible. Yet it captures the essence of what makes machine learning revolutionary: computers figuring things out on their own.
Tom Mitchell (1997)
Fast forward to 1997, when Tom Mitchell offered a more formal machine learning definition:
“A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.”
Mitchell’s definition is particularly loved by ML students for its precision. It breaks learning down into three components: the experience (data), the task (what we want the computer to do), and the performance measure (how well it’s doing it).
Modern Machine Learning Definitions From Leading Books
As machine learning has exploded in popularity, authors of influential books have contributed their own machine learning definitions, each adding unique perspective.
From “Modelling Mindsets” by Christoph Molnar
Molnar defines machine learning as:
“A branch of artificial intelligence that deals with improving at a given task through experience, which means learning from data.”
What’s interesting here is the explicit connection to artificial intelligence and the emphasis on experience. Molnar’s definition highlights machine learning’s roots in AI while emphasizing the critical role of learning from data.
From “Python Machine Learning” by Sebastian Raschka
In Python Machine Learning Sebastian Raschka takes a more computer science-oriented approach to his definition:
“Machine learning is a subfield of computer science that is concerned with building algorithms which, to be useful, rely on a collection of examples of some phenomenon.”
Raschka’s definition emphasizes the algorithmic nature of machine learning and the fundamental importance of examples. This focus on examples highlights the data-driven nature of machine learning as opposed to rule-based programming.
From “Hands-On Machine Learning” by François Chollet
In his most famous book, the creator of Keras, a popular deep learning library, offers this definition:
“Machine Learning is the science (and art) of programming computers so they can learn from data.”
What makes Chollet’s definition stand out among other definitions is the inclusion of art alongside science. This acknowledges that building effective machine learning systems involves creativity and intuition, not just technical knowledge.
Deep Learning Experts Weigh In
As deep learning has revolutionized machine learning, its pioneers have contributed their own definitions.
From “Deep Learning” by Goodfellow, Bengio & Courville
In their cornerstone book about Deep Learning, these giants in the field define machine learning as:
“Machine learning allows computers to extract patterns from data and use them to make predictions or decisions, without being explicitly told what to do.”
This definition emphasizes pattern recognition and autonomous decision-making, key aspects of modern machine learning systems. The focus on extraction highlights how machine learning algorithms discover knowledge hidden within data.
From “Machine Learning Yearning” by Andrew Ng
Andrew Ng, who has done perhaps more than anyone to popularize machine learning education, offers this definition:
“Machine learning is the science of getting computers to learn without being explicitly programmed.”
While similar to Samuel’s pioneering definition, Ng’s carries the weight of coming from someone who helped usher in the deep learning revolution that has transformed the field.
The Common Threads in Machine Learning Definitions
Looking across all these definitions from various experts and eras, we can identify common threads that define the field:
- Learning from data/experience rather than explicit programming
- Improvement over time
- Pattern recognition and extraction
- Making predictions or decisions
- Generalization to new situations
Different experts emphasize different aspects, but these core ideas appear consistently across definitions through the decades.
Conclusion on Machine Learning Definitions
ML definitions have evolved alongside the technology itself, but the fundamental concept remains consistent: systems that improve through experience rather than explicit programming. From Samuel’s pioneering definition to modern characterizations by deep learning experts, each perspective adds richness to our understanding.
Oh, and by the way, which definition resonates most with you? Is it Mitchell’s precise formula? Chollet’s inclusion of artistry? Or perhaps Samuel’s remarkably prescient characterization from the dawn of computing? Whatever your preference, these machine learning definitions collectively help us grasp this transformative technology.
If you want to learn more about how machine learning is changing our world, check out our follow-up articles on practical applications and ethical considerations in AI!
As always, thank you for reading How to Learn Machine Learning, and have a wonderful day!
Subscribe to our awesome newsletter to get the best content on your journey to learn Machine Learning, including some exclusive free goodies!
