Machine learning is a subset of artificial intelligence that gives systems the ability to learn and optimize processes without having to be consistently programmed. Simply put, machine learning uses data, statistics and trial and error to “learn” a specific task without ever having to be specifically coded for the task.
What Is Machine Learning?
Machine learning is a subfield of artificial intelligence in which systems have the ability to “learn” through data, statistics and trial and error in order to optimize processes and innovate at quicker rates. Machine learning gives computers the ability to develop human-like learning capabilities, which allows them to solve some of the world’s toughest problems, ranging from cancer research to climate change.
Most computer programs rely on code to tell them what to execute or what information to retain (better known as explicit knowledge). This knowledge contains anything that is easily written or recorded, like textbooks, videos or manuals. With machine learning, computers gain tacit knowledge, or the knowledge we gain from personal experience and context. This type of knowledge is hard to transfer from one person to the next via written or verbal communication.
Facial recognition is a type of tacit knowledge. We recognize a person’s face, but it is hard for us to accurately describe how or why we recognize it. We rely on our personal knowledge banks to connect the dots and immediately recognize a person based on their face. Another example is riding a bike. It’s much easier to show someone how to ride a bike than it is to explain it.
Computers no longer have to rely on billions of lines of code to carry out calculations. Machine learning gives computers the power of tacit knowledge that allows these machines to make connections, discover patterns and make predictions based on what it learned in the past. Machine learning’s use of tacit knowledge has made it a go-to technology for almost every industry from fintech to weather and government.
How Does Machine Learning Work?
Machine learning compiles input data, which can be data gathered from training sessions or other sources, such as data set search engines, .gov websites and open data registries like that of Amazon Web Services. This data serves the same function that prior experiences do for humans, giving machine learning models historical information to work with when making future determinations.
Algorithms then analyze this data, searching for patterns and trends that allow them to make accurate predictions. In this way, machine learning can glean insights from the past to anticipate future happenings. Typically, the larger the data set that a team can feed to machine learning software, the more accurate the predictions.
The idea is that machine learning algorithms should be able to perform these tasks on their own, requiring minimal human intervention. This speeds up various processes as machine learning comes to automate many aspects of different industries.
Types of Machine Learning
Like all systems with AI, machine learning needs different methods to establish parameters, actions and end values. Machine learning-enabled programs come in various types that explore different options and evaluate different factors. There is a range of machine learning types that vary based on several factors like data size and diversity. Below are a few of the most common types of machine learning under which popular machine learning algorithms can be categorized.
Supervised Learning
Supervised learning involves mathematical models of data that contain both input and output information. Machine learning computer programs are constantly fed these models, so the programs can eventually predict outputs based on a new set of inputs.
Regression and classification are two of the more popular analyses under supervised learning. Regression analysis is used to discover and predict relationships between outcome variables and one or more independent variables. Commonly known as linear regression, this method provides training data to help systems with predicting and forecasting. Classification is used to train systems on identifying an object and placing it in a sub-category. For instance, email filters use machine learning to automate incoming email flows for primary, promotion and spam inboxes.
Unsupervised Learning
Unsupervised learning contains data only containing inputs and then adds structure to the data in the form of clustering or grouping. The method learns from previous test data that hasn’t been labeled or categorized and will then group the raw data based on commonalities (or lack thereof). Cluster analysis uses unsupervised learning to sort through giant lakes of raw data to group certain data points together. Clustering is a popular tool for data mining, and it is used in everything from genetic research to creating virtual social media communities with like-minded individuals.
Semi-Supervised Learning
Semi-supervised learning falls in between unsupervised and supervised learning. Instead of giving a program all labeled data (like in supervised learning) or no labeled data (like in unsupervised learning), these programs are fed a mixture of data that not only speeds up the machine learning process, but helps machines identify objects and learn with increased accuracy.
Typically, programmers introduce a small number of labeled data with a large percentage of unlabeled information, and the computer will have to use the groups of structured data to cluster the rest of the information. Labeling supervised data is seen as a massive undertaking because of high costs and hundreds of hours spent.
Here is a handy way to remember the differences between machine learning types: Supervised learning is like being a student and having the teacher constantly watch over you at school and at home. Unsupervised learning is telling a student to figure a concept out themselves. Semi-supervised learning is like giving a student a lesson and then testing them on questions pertinent to that topic. Each machine learning type has its advantages and disadvantages, and all are used based on the parameters and needs of the data scientist or engineer.
Machine Learning Examples and Applications
Financial Services
The financial services industry is championing machine learning for its unique ability to speed up processes with a high rate of accuracy and success. What has taken humans hours, days or even weeks to accomplish can now be executed in minutes. There were over 581 billion transactions processed in 2021 on card brands like American Express. Ensuring these transactions are more secure, American Express has embraced machine learning to detect fraud and other digital threats.
Additionally, machine learning is used by lending and credit card companies to manage and predict risk. These computer programs take into account a loan seeker’s past credit history, along with thousands of other data points like cell phone and rent payments, to deem the risk of the lending company. By taking other data points into account, lenders can offer loans to a much wider array of individuals who couldn’t get loans with traditional methods.
Trading firms are using machine learning to amass a huge lake of data and determine the optimal price points to execute trades. These complex high-frequency trading algorithms take thousands, if not millions, of financial data points into account to buy and sell shares at the right moment.
Healthcare
The healthcare industry uses machine learning to manage medical information, discover new treatments and even detect and predict disease. Medical professionals, equipped with machine learning computer systems, have the ability to easily view patient medical records without having to dig through files or have chains of communication with other areas of the hospital. Updated medical systems can now pull up pertinent health information on each patient in the blink of an eye.
AI and machine learning can automate maintaining health records, following up with patients and authorizing insurance — tasks that make up 30 percent of healthcare costs.
Machine learning-enabled AI tools are working alongside drug developers to generate drug treatments at faster rates than ever before. Essentially, these machine learning tools are fed millions of data points, and they configure them in ways that help researchers view what compounds are successful and what aren’t. Instead of spending millions of human hours on each trial, machine learning technologies can produce successful drug compounds in weeks or months.
Machine learning has made disease detection and prediction much more accurate and swift. Machine learning is employed by radiology and pathology departments all over the world to analyze CT and X-RAY scans and find disease. After being fed thousands of images of disease through a mixture of supervised, unsupervised or semi-supervised models, some machine learning systems are so advanced that they can catch and diagnose diseases (like cancer or viruses) at higher rates than humans. Machine learning has also been used to predict deadly viruses, like Ebola and Malaria, and is used by the CDC to track instances of the flu virus every year.
Social Media
Machine learning is employed by social media companies for two main reasons: to create a sense of community and to weed out bad actors and malicious information. Machine learning fosters the former by looking at pages, tweets, topics and other features that an individual likes and suggesting other topics or community pages based on those likes. It’s essentially using your preferences as a way to power a social media recommendation engine.
The spread of misinformation in politics has prompted social media companies to use machine learning to quickly identify harmful patterns of false information, flag malicious bots, view reported content and delete when necessary in order to build online communities based on truth.
Retail and E-commerce
The retail industry relies on machine learning for its ability to optimize sales and gather data on individualized shopping preferences. Machine learning offers retailers and online stores the ability to make purchase suggestions based on a user’s clicks, likes and past purchases. Once customers feel like retailers understand their needs, they are less likely to stray away from that company and will purchase more items.
Visual search is becoming a huge part of the shopping experience. Instead of typing in queries, customers can now upload an image to show the computer exactly what they’re looking for. Machine learning will analyze the image (using layering) and will produce search results based on its findings.
Machine learning has also been an asset in predicting customer trends and behaviors. These machines look holistically at individual purchases to determine what types of items are selling and what items will be selling in the future. For example, maybe a new food has been deemed a “super food.” A grocery store’s systems might identify increased purchases of that product and could send customers coupons or targeted advertisements for all variations of that item. Additionally, a system could look at individual purchases to send you future coupons.
What Is Deep Learning?
Deep learning is a subfield within machine learning, and it’s gaining traction for its ability to extract features from data. Deep learning uses Artificial Neural Networks (ANNs) to extract higher-level features from raw data. ANNs, though much different from human brains, were inspired by the way humans biologically process information. The learning a computer does is considered “deep” because the networks use layering to learn from, and interpret, raw information.
For example, deep learning is an important asset for image processing in everything from e-commerce to medical imagery. Google is equipping its programs with deep learning to discover patterns in images in order to display the correct image for whatever you search. If you search for a winter jacket, Google’s machine and deep learning will team up to discover patterns in images — sizes, colors, shapes, relevant brand titles — that display pertinent jackets that satisfy your query.
Deep learning is also making headwinds in radiology, pathology and any medical sector that relies heavily on imagery. The technology relies on its tacit knowledge — from studying millions of other scans — to immediately recognize disease or injury, saving doctors and hospitals both time and money.
History of Machine Learning
Machine learning has been a field decades in the making, as scientists and professionals have sought to instill human-based learning methods in technology.
The brief timeline below tracks the development of machine learning from its beginnings in the 1950s to its maturation during the twenty-first century.
1950
Alan Turing jumpstarts the debate around whether computers possess artificial intelligence in what is known today as the Turing Test. The test consists of three terminals — a computer-operated one and two human-operated ones. The goal is for the computer to trick a human interviewer into thinking it is also human by mimicking human responses to questions.
1956
Arthur Samuel publicly reveals a computer that can determine the optimal moves to make in a checker match.
1957
Frank Rosenblatt creates the first neural network for computers, known as the perceptron. This invention enables computers to reproduce human ways of thinking, forming original ideas on their own.
1962
Samuel builds on previous versions of his checkers program, leading to an advanced system made for the IBM 7094 computer. In 1962, the computer defeats checkers master Robert Nealy in a match.
1979
Descending from a line of robots designed for lunar missions, the Stanford cart emerges in an autonomous format in 1979. The machine relies on 3D vision and pauses after each meter of movement to process its surroundings. Without any human help, this robot successfully navigates a chair-filled room to cover 20 meters in five hours.
1981
Gerald Dejong explores the concept of explanation-based learning (EBL). This approach involves providing a computer with training data, which it analyzes to develop a rule for filtering out unnecessary information. The idea is that this data is to a computer what prior experience is to a human being.
1985
Researcher Terry Sejnowksi creates an artificial neural network of 300 neurons and 18,000 synapses. Called NetTalk, the program babbles like a baby when receiving a list of English words, but can more clearly pronounce thousands of words with long-term training.
1990s
The 1990s marks a shift in the realm of machine learning. Scientists focus less on knowledge and more on data, building computers that can glean insights from larger data sets.
1997
Scientists at IBM develop a computer called Deep Blue that excels at making chess calculations. The program defeats world chess champion Garry Kasparov over a six-match showdown.
2006
“Deep learning” becomes a term coined by Geoffrey Hinton, a long-time computer scientist and researcher in the field of AI. He applies the term to the algorithms that enable computers to recognize specific objects when analyzing text and images.
2010
Microsoft releases a motion-sensing device called Kinect for the Xbox 360. The device contains cameras and sensors that allow it to recognize faces, voices and movements. As a result, Kinect removes the need for physical controllers since players become the controllers.
2011
IBM’s Watson competes on Jeopardy! against two of the show’s most decorated champions. That same year, Google develops Google Brain, which earns a reputation for the categorization capabilities of its deep neural networks.
2012
Computer scientists at Google’s X lab design an artificial brain featuring a neural network of 16,000 computer processors. The network applies a machine learning algorithm to scan YouTube videos on its own, picking out the ones that contain content related to cats.
2014
Facebook unveils its new face recognition tool DeepFace. Composed of a deep network of millions of data points, DeepFace leverages 3D face modeling to recognize faces in images in a way very similar to that of humans.
2015
Amazon develops a machine learning platform while Microsoft releases its Distributed Machine Learning Toolkit. In response to the proliferation of machine learning and AI, over 3,000 AI and robotics researchers — including names like Stephen Hawking and Elon Musk — sign an open letter warning of AI-powered warfare.
2017
Google’s AI algorithm AlphaGo specializes in the complex Chinese board game Go. The algorithm achieves a close victory against the game’s top player Ke Jie in 2017. This win comes a year after AlphaGo defeated grandmaster Lee Se-Dol, taking four out of the five games.
