قالب وردپرس درنا توس
Home / Technology / Google (GOOGL) has the most important artificial intelligence edge over Microsoft (MSFT)

Google (GOOGL) has the most important artificial intelligence edge over Microsoft (MSFT)



Microsoft (MSFT) and Alphabet (GOOGL) are honestly similar stocks, which have fallen exactly the same amount this year, have similar growth profiles and identical market values ​​and valuations.

Which ones do you choose?

Both are good businesses, and a credible case can be made for either. Removing the differences in their revenue lines – operating system software versus advertising, for example – comes down to dominance in the software ecosystem. And in this respect, both Microsoft and Alphabet have powerful cloud computing operations that each provide leadership in the most important category of computer software today, machine learning.

But the alphabet's Google device has no doubt whatsoever on Microsoft in any way.

At $ 751

billion in market value, Microsoft topped Google's market value of $ 707 billion, back in September this year, and held a 13% gain for the year while the alphabet is around 4%. The forward-to-earnings value for the two is equal, while Microsoft is slightly cheaper, 19 times versus the alphabet 22 times. Both are quite expensive compared to the general market, of course. Google has the edge on revenue growth, with 19% projected for 2019, against 11% for Microsoft's fiscal 2020 revenue growth.

The difference between the two comes down to what happens in these cloud operations. Microsoft has a scale, but it cannot have the more important relative position with developers.

Machine learning, whereby computers can develop some of the code they use automatically by extracting patterns from data, is the most important trend in programming today. [19659002] Both companies, along with Amazon (AMZN), have implemented machine learning tools in their cloud computing operations. Microsoft's Azure shooter service is generally the only other one for Amazon Web Services in terms of revenue, with Google Cloud one third. But since neither Microsoft nor Google reveals revenue for the devices, the comparison is sketchy.

However, it is safe to say that cloud operations are important assets, where both companies attract and retain developers of the machine learning network.

However, Google has control over the key framework for building machine learning programs, called TensorFlow. A framework is a collection of pre-built code that can be used by a software developer to collect applications without reinventing the wheel. Right now, TensorFlow is the dominant framework for machine learning.

It's not the only one, and actually Microsoft has some important deals on its own. They include something called the "infer.NET" frame, and another called "Microsoft Cognitive Toolkit." Both are used by developers in a number of interesting projects.

But over and over, the large amount of work that comes into machine learning comes through TensorFlow; either that or another pair of companies other than Microsoft, such as "PyTorch" from Facebook; a framework called "Caffe" originally developed by the University of California at Berkley; and recent initiatives based on these programs, such as "fast.ai" at the San Francisco startup of that name.

Nothing about Microsoft's $ 7.5 billion acquisition of developer GitHub in October is changing some of it. Machine learning frameworks gain on the strengths or weaknesses they possess, as well as how much they are supported. Owning the most popular programming directory, GitHub, does not change the frameworks decisions that some developers make.

Why framework control is that Google has built on the use of TensorFlow. They have their own custom chips in their cloud operation, called "Tensor Processing Units" or TPUs, which have become a leading piece of silicon and are made available by Google in its cloud computing service. Combined with TensorFlow software, TPUer Google provides a chip and software combination that is an edge over what Microsoft currently offers.

And TensorFlow has spread rapidly to mobile computing, where Google has the dominant worldwide presence, through the control of the Android operating system, and Microsoft has no presence. A study released in November by scientists including Peking University, but also Microsoft's own research unit in Beijing, found that among the best mobile applications for Android, by those who use machine learning in some form, use half as much as either TensorFlow; "TFLite", a minor subset of TensorFlow created by Google only for mobile devices; or "ncnn", another framework created by Chinese social networking service Tencent (TCEHY). Microsoft's code isn't even started as far as machine learning on mobile devices.

This whole field of machine learning is still very young and very new. There is a lot of innovation to come, and Google's relative leads to shrink and evaporate over time. An interesting initiative from Microsoft is its struggle for something called "Open Neural Network Exchange" or ONNX, which can make developers less dependent on individual frames over time.

Time will tell: Meanwhile, Google's alphabet has the edge of Microsoft in the core tools used for the most groundbreaking types of application development. It's worth a bid.

Alphabet, Microsoft and Amazon are holdings in Jim Cramer's Action Alerts PLUS Member Club . Want to be notified before Jim Cramer buys or sells GOOGL, MSFT or AMZN? Learn more now.


Source link