Machine learning, a branch of artificial intelligence, concerns the construction and study of systems that can learn from data. For example, a machine learning system could be trained on email messages to learn to distinguish between spam and non-spam messages. After learning, it can then be used to classify new email messages into spam and non-spam folders.
The core of machine learning deals with representation and generalization. Representation of data instances and functions evaluated on these instances are part of all machine learning systems. Generalization is the property that the system will perform well on unseen data instances; the conditions under which this can be guaranteed are a key object of study in the subfield of computational learning theory.
Machine Learning can also be described as optimizing a performance criterion using example data and past experience.In machine learning, data plays an indispensable role, and the learning algorithm is used to discover and learn knowledge or properties from the data. The quality or
quantity of the dataset will affect the learning and prediction performance.
In daily life, people are easily facing some decisions to make. For example, if the sky is cloudy, we may decide to bring an umbrella or not. For a machine to make these kinds of choices, the intuitive way is to model the problem into a mathematical expression. The mathematical expression could directly be designed from the problem background. For instance, the vending machine could use the standards and security decorations of currency to detect false money. While in some other problems that we can only acquire several measurements and the corresponding labels, but do not know the specific relationship among them, learning will be a better way to find the underlying connection.
Another great illustration to distinguish designing from learning is the image compression technique. JPEG, the most widely used image compression standard, exploits the block-based DCT to extract the spatial frequencies and then unequally quantizes each frequency component to achieve data compression. The success of using DCT comes from not only the image properties, but also the human visual perception. While without counting the side information, the KL transform
(Karhunen-Loeve transform), which learns the best projection basis for a given image, has been proved to best reduce the redundancy . In many literatures, the knowledge acquired from human understandings or the intrinsic factors of problems are called the domain knowledge. And the knowledge learned from a given training set is called the data-driven knowledge.
There are generally three types of machine learning based on the ongoing problem and the given data set, (1) supervised learning, (2) unsupervised learning, and (3) reinforcement learning:
Supervised learning tries to find the relationships between the feature set and the label set, which is the knowledge and properties we can learn from labeled dataset.The knowledge extracted from supervised learning is often utilized for prediction and recognition.
Unsupervised learning aims at clustering , probability density estimation, finding association among
features, and dimensionality reduction. In general, an unsupervised algorithm may simultaneously learn more than one properties, and the results from unsupervised learning could be further used for supervised learning.
Reinforcement learning is used to solve problems of decision making (usually a sequence of decisions), such as robot perception and movement, automatic chess player, and automatic vehicle driving
Machine learning will enable cognitive systems to learn, reason and engage with us in a more natural and personalized way. These systems will get smarter and more customized through interactions with data, devices and people. They will help us take on what may have been seen as unsolvable problems by using all the information that surrounds us and bringing the right insight or suggestion to our fingertips right when it’s most needed. Over the next five years, machine learning applications will lead to new breakthroughs that will amplify human abilities, assist us in making good choices, look out for us and help us navigate our world in powerful new ways.
Some applications of machine learning are as follows:
The classroom will learn you
The classroom of the future will learn about each student over the course of their education, helping students master the skills critical to meeting their goals. A system fueled by sophisticated analytics over the cloud will help teachers predict students who are most at risk, their roadblocks, and then suggest measures to help students overcome their challenges.
Buying local will beat online
Buying local will be back in style once again. Savvy retailers will use the immediacy of the physical store and proximity to customers to create experiences that cannot be replicated by online-only retail. They will magnify the digital experience by bringing the web right to where the shopper can physically touch it.
Doctors will routinely use your DNA to keep you well
Computers will help doctors understand how a tumor affects a patient down to their DNA, and present a collective set of medications shown to best attack the cancer, while reducing the time it takes to find the right treatment for a patient from weeks and months to days and minutes.
A digital guardian will protect you online
Security is going to become more agile and contextual based on a 360 degree of an individual’s data, devices and applications. A digital guardian will have your back, trained to focus on the people and items it is entrusted with, so it can make inferences about what’s normal or reasonable activity and what’s not, ready to spot deviations that could be precursors to an attack and a stolen identity.
The city will help you live in it
Smarter cities will become sentient cities, understanding in real time how billions of events occur as computers learn to understand what people need, what they like, what they do, and how they move from place to place. Mobile devices and social engagement will enable citizens to strike up a relationship with their city leaders so their voices will be heard not only on election day, but every day.
There are many other applications of machine learning, As machines become smarter, they will aid humans in their quest for the supreme.