In machine learning, the goal is to solve some complex computational task by “letting the machine learn”.
Programmed by Arthur Samuel, this big data discipline of artificial intelligence replaces the tedious task of trying to understand the problem well enough to be able to write a program, which can take much longer or be virtually impossible.
Techopedia defines the discipline of machine learning as “an artificial intelligence (AI) discipline geared toward the technological development of human knowledge. Machine learning allows computers to handle new situations via analysis, self-training, observation and experience. Machine learning facilitates the continuous advancement of computing through exposure to new scenarios, testing and adaptation, while employing pattern and trend detection for improved decisions in subsequent (though not identical) situations.”
Thank Arthur Samuel and the game of checkers for machine learning.
In 1959, IBM employee Arthur Samuel wanted to teach a computer to play checkers. So, he wrote the original program on IBM’s first commercial computer, the IBM 701, but he kept winning. (Samuel defines it as a “field of study that gives computers the ability to learn without being explicitly programmed”).
Hence, Samuel wrote a program to let the computer play against itself. The program collected data on its games and created a predictive engine, to optimize its playing and Samuel started losing. Ultimately, the program was a success; so much that IBM stock went up 15 points in one day when it was announced!
As an aside, artificial intelligence has a longer history. In Greek mythology, Hephaestus, the Greek god of technology, built intelligent robots. Acting as if they were alive, and often appearing to be, except for the fact that they exist out of metal. Talos, one of his creations, was built to guard the island of Crete!
Recently, we see a number of different industries incorporating machine learning, as the table below illustrates. These are all outside of IT.
Some of the categories of machine learning algorithms are: Supervised algorithms, Unsupervised algorithms, and Reinforcement algorithms.
Machine learning does not equal “statistics”. However, it will use such techniques as Naïve Bayes Classifier Algorithm, K Means Clustering Algorithm, Support Vector Machine Algorithm, Linear Regression, and Logistic Regression. These go a bit beyond Stats 101!
Furthermore, machine learning is NOT rules-based. Rules are based on what a human knows about the data. When not tuned properly, rules generate excessive alerts. These focus on known unknowns; but what about unknown unknowns? Humans cannot predict what future attacks will look like.
”Something like IoT just would not be possible without machine learning because there is just too much data coming from all sorts of devices,” – Nick Patience, 451 Research
Finally, machine learning uses automated and iterative algorithms, such as the above, to learn about patterns in big data, detecting anomalies, and identifying a structure that may be new and previously unknown. Therefore, when paired with statistical analysis, it identifies relationships that may otherwise have gone undetected. In conclusion, it can surpass human capability and software engineering capability to make use of volumes of big data.
Some of the major benefits of machine learning:
- Accelerated time in model development and delivery of actionable insights
- An optimal balance between predictive accuracy, performance and cost
- Utilization of streaming data to deliver real-time analysis
- Reduction of risk with enterprise grade machine learning
- Acquisition of best insights in model performance and outcome
Global organizations use Gurucul technology to detect insider threats, cyber fraud, IP theft, external attacks and more. The company is based in Los Angeles. To learn more, visit Gurucul.com and follow us on LinkedIn and Twitter.