The buzzword bingo of AI and Machine Learning

by | Jun 14, 2017

Like Cyber Security both Artificial Intelligence and Machine Learning are well and truly in their hype cycle right now and quite rightly so. No longer the sole property of science fiction or future dystopian society novels our approach to building applications, leveraging data, solving problems, running a business, trading stocks, treating patients, legal proceedings, insurance <insert any industry here> is changing as a result of these technology advances.
When anything enters it’s hype cycle we also enter a phase where vendors make grander and grander claims their product is the best of the best. With this comes buzzword bingo a familiar period where words are repurposed and occasionally invented to support how unique and special the hyped product can be.
Cynicism of hype cycles aside – Artificial Intelligence (AI) and Machine Learning (ML) are here to change our lives. Before I start on clarifying terms and introducing who the players are it’s worthy noting that AI and ML are both commonly used as umbrella terms, both of which are morphing as our use and understanding of technology and methodologies grow.

Clarifying the Terms

There are many definitions, good old Wikipedia ones, commentator ones and vendor-led ones. I do like the Analogies some of those pose. Forrester researched whether the AI hype was real (only 12% of companies surveyed late 2016 had employed AI techniques) and within their assessment provide this definition of AI:

Enterprises that plan to invest in AI expect to improve customer experiences, improve products and services, and disrupt their industry with new business models.

  • “Artificial” is the opposite of organic. Artificial simply means person-made versus occurring naturally in the universe. Computer scientists, engineers, and developers research, design, and create a combination of software, computers, and machine to manifest AI technology.
  • “Intelligence” is in the eye of the beholder.  Philosophers will have job security for a very long time trying to define intelligence precisely. That’s because, intelligence is much tougher to define because we humans routinely assign intelligence to all matter of things including well-trained dachshunds, self-driving cars, and “intelligent” assistants such as Amazon Echo. Intelligence is relative. For AI purists, intelligence is more akin to human abilities. It means the ability to perceive its environment, take actions that satisfy a set of goals, and learn from both successes and failures. Intelligence among humans varies greatly and so too does it vary among AI systems.

I looked to IBM for similar insight into AI and got lost in the IBM Watson world but did find a nifty definition of ML complete with examples:

Machine Learning is the 21st century’s Industrial Revolution

  • Machine learning enables cognitive systems to learn, reason and engage with us in a natural and personalized way. Think Netflix movie recommendations, Internet ads based on browsing habits, or even stock trades — these are all ways machine learning is helping us navigate our world in powerful new ways.

Taking the example approach a step further Maavdnbo describe “Artificial Intelligence is human intelligence exhibited by machines. Machine Learning is an approach to achieve artificial intelligence. Deep Learning is a technique for implementing machine learning.” Then go on to say:

Artificial Intelligence: The art of beating a human at chess

  • Ask expert chess players for their strategy and implement it as a combination of ‘If this then that’ rules.
  • Gather data on every chess game between two humans. For every situation, plan out the possible actions and the probability of winning the game for each action. Let the system always choose the action that gives the highest probability of winning the game.

Machine Learning: The art of beating a human at chess again and again

  • Let the computer play millions of games and gather data on winning probabilities for every action in every possible situation. Make it constantly learn to adjust its own choices, including as many hand-written parameters you can imagine.

Who are the players

There are so many use cases for both AI and ML and a range of ways to represent who the players in this market are. Couple this with a constantly evolving landscape in a relatively early stage evolution I found there are too many to describe in one blog. So here is are links to many of the lists out there, most include technology platforms and building block players as well as specialist providers in this space:

You will have noticed AWS, Microsoft Azure, Google Deepmind, IBM, Apple, Facebook, OpenAI all seem to be dominant names alongside a nice list of Opensource software and range of niche players.
On that note, Shaun from our OptimalBI team wrote a nifty comparison of how then new-on-the-scene AWS and more mature provider Microsoft have approached Machine Learning back in 2015 that is also worth a read.

What are the buzzwords

If you want to play in this space you need a basic understanding of the lingo, you will notice many of these aren’t new terms or exclusive to the AI / ML vocabulary and many are tried and true.
The NZ Government have put together a buzzword filled Infographic if you would like to see a visual representation – Callaghan Innovation AI Infographic. Otherwise here is a list of terms injected into AI and ML language with links to wikipedia definitions:

Watch this space

As I opened with, Artificial Intelligence and Machine Learning are in their hype cycle, if you are in the software game or Internet of Things (IoT) device space you should already be embracing these technologies and methods. This arena is moving quite quickly so it’s going to be interesting to see the evolution unfold, with companies like IBM backing their entire future on investment here there is certainly significant investment happening.
Even reaching little of NZ, a collection of companies have come together and established a forum to focus our conversations and find opportunities to collaborate which is great news.
Where the real effort needs to go next is into legislation, ethics and behaviour – a blog post for another day – so as ever I will leave the last word to Dilbert (thank you Vic.

Have you seen our great TEC AgileBI Coaching Case Study– great insight into the benefits of Agile Business Intelligence projects from our customer. Enjoy. Vic. 

Submit a Comment

Your email address will not be published. Required fields are marked *