5 Top Machine Learning Stocks to Buy Now

Machine learning is a budding branch of artificial intelligence (AI) that has become all the rage, and industry after industry is climbing on board this ongoing trend. The biggest and brightest minds have become increasingly enthusiastic about the societal and business applications for this emerging technology. Some of the stock market's biggest companies are using the technology to improve performance and gain a competitive edge.

While the market potential for machine learning is massive, there is no consensus on just how big it could eventually be. The global machine learning market is forecast to grow to $8.81 billion in 2022, producing a compound annual growth rate of 44%, according to a report by MarketsandMarkets Research. Deep learning, one technique of machine learning, was worth an estimated $4.8 billion in 2017, and will skyrocket to $261 billion by 2027, producing annual growth of 49%, according to a report by Persistence Market Research. While the exact numbers are elusive, the potential is impressive.

Machine learning has sparked an AI revolution. Image source: Getty Images.

The opportunity and its tremendous prospects have not gone unnoticed, and a growing number of companies are enlisting the technology in an ever-expanding array of use cases. This list represents just a few of the more prominent players.

Data source: Yahoo! Finance. Chart by author.

While the choices for investment in this area cover a wide area, a few companies represent what I believe are particularly compelling opportunities. Read on to see why Netflix, NVIDIA, Amazon, Alphabet, and Apple are my top picks in the realm of machine learning.

What is machine learning?

Before delving into the potential winners in the space, it may be helpful to first define this breakthrough technology. Machine learning is a branch of artificial intelligence that gives computers the ability to learn without being explicitly programmed to do so. Machine learning systems use algorithms to sift through large amounts of data, find patterns among the many data points that might otherwise be overlooked, and learn from what it finds. The system can independently adapt, improve, and make predictions or determinations based on what it finds.

If you think this definition sounds very similar to deep learning -- you're right. Deep learning is one of numerous techniques used to train machine-learning models.

Presented for your viewing pleasure

Video-streaming pioneer Netflix may not seem like an obvious choice when it comes to machine learning, but this would ignore the fact that the company uses machine learning algorithms to inform many of its business decisions. Netflix has more than 20 years of data from its subscribers that it uses to power its recommendation engine, but that's just the beginning.

Netflix famously commissioned its original series House of Cards based on algorithms that indicated the show would be a hit. Machine learning also helps personalize the artwork each member sees to highlight a given title. The company uses machine learning to resize video images, to tailor the compression of each frame to match the available bandwidth in the area, which prevents buffering for mobile users. Netflix even uses algorithms to provide more effective advertising.

Those moves are paying off. In its recent quarter, Netflix grew domestic subscribers by 10.7% year over year, while international subscribers jumped 41.6%. Numbers like that show that its algorithms are working.

In an odd bit of art imitating life, AI will figure prominently in Netflix's original thriller Altered Carbon, where the technology allows the rich in a futuristic society to download their consciousness and memories into a new body.

Algorithms have paved the way for advances in machine learning. Image source: Getty Images.

A chip off the old block

NVIDIA was at the forefront of the AI revolution. It took a combination of big data, the right algorithms, and lightning-fast processors to make machine learning work -- and NVIDIA graphic processing units (GPUs) provided the final piece. Parallel processing, the ability to perform a multitude of complex mathematical calculations simultaneously, worked just as well in machine learning applications as it did in rendering graphics.

NVIDIA quickly became the go-to for the majority of companies seeking to leverage machine learning. That worked its way into the company's financial results and its stock price, which has grown 1,000% in the last three years alone. The company's data center revenue, which is derived from machine learning, has grown more than 100% year over year in each of the last six quarters, and it now represents 19% of NVIDIA's total sales.

Head in the clouds

E-commerce leader Amazon has been using machine learning for 20 years to forecast inventory, manage its fulfillment centers, recommend products, and provide better search results. 

More recently, Amazon infused its machine learning into a small cylinder, and the Echo smart speaker was born. While Amazon isn't talking, the devices have made their way into an estimated 20 million U.S. homes -- accounting for 73% of the smart speaker market, according to data from Consumer Intelligence Research Partners (CIRP). Even more importantly for Amazon, consumers that own the devices spend 66% more than the average customer. 

The company has also infused machine learning into its highly profitable Amazon Web Services (AWS) cloud computing service, making these tools available to its customers. Moves like this continue to drive growth at AWS. In its 2017 third quarter, the cloud computing segment generated 10% of Amazon's revenue and all of its profits -- even subsidizing the growth of its money-losing e-commerce segment. During the quarter, AWS generated $4.6 billion in sales, up 55% year over year, and contributed $1.17 billion in operating profit.

Machine learning can process reams of data and find hidden patterns. Image source: Getty Images.

Now I know my ABCs...

Alphabet is best known for its Google search, but the company was one of the early adopters of machine learning and has long used algorithms to improve its search results. Over the last couple of years, Google has been quietly integrating machine learning into many of its headline features, including Image Search, Translate, and Photos. Smart Reply, which suggests potential replies to emails and texts, is also a result of the technology.

Google's focus on machine learning is paying off in more tangible ways as well. The company applied machine learning to understand how to more efficiently arrange its massive server farms. Its machine learning system was able to realize a 40% reduction in the amount of energy used to cool its data centers, which amounted a 15% increase in power usage efficiency. This saved the company hundreds of millions of dollars. 

Internally, the company began a program that embeds engineers in a machine learning team for six-month stints. Googlers then take this new-found knowledge and find ways to improve processes in their own areas. Only an estimated 10% of Google's 25,000 engineers are proficient in machine learning, but the company would like it to be closer to 100%. It's been teaching in-house classes on the subject since 2005. 

An Apple a day

Apple is another company that may not immediately spring to mind as a leader in machine learning. After an initial splash with its Siri voice-activated digital assistant, some thought Apple was being left behind by more technically adept Silicon Valley competitors. Many believed the company's focus on privacy was causing it to fall behind, and its refusal to let its scientists publish work in the field didn't help with that perception.

That view began to change when Apple decided in late 2016 that it would break from its long-standing practice and allow its researchers to begin publishing their work. Apple provided further evidence of its advances in the field with the release of the iPhone 8 and iPhone X. Apple brought machine learning from the cloud to the phone and integrated the technology into many of the iPhone's groundbreaking features like Face ID, intelligent camera sensors, and augmented reality features. These moves show that Apple is leading the field in integrating machine learning into its flagship product.

Show me the money

Not every use of machine learning will draw a straight line to a company's financial statements. While the sales of NVIDIA's GPUs for machine learning can simply be counted, increases in customer satisfaction for Google or Netflix aren't as easy to quantify.

The inability to see how AI will translate to the bottom line shouldn't stop investors from understanding the transformational nature of machine learning and the near limitless applications. These companies have an early lead in developing ways to profit from the nascent technology and present compelling opportunities for investors.