Artificial intelligence (AI) is an oft misunderstood umbrella term. Technologies like machine learning (ML) and cognitive computing come under the AI umbrella, but the terms are not synonymous. Christopher Glode, Chief Digital Officer of WellBiz Brands, notes, “Anywhere you find an executive talking about technology trends, you’ll almost certainly hear ‘AI and ML’ as a transformational technology that has the potential to radically change their business. This may be true, but when pressed for details, these executives will admit they have only a nominal understanding of the nuts and bolts of how machine learning actually works.”[1] He believes this lack of understanding is a big problem. He continues, “If you’re an executive claiming to use ML to enhance your business, you ought to be able to confidently answer these questions: What task are you addressing or automating, and what problem are you solving? What training data are you using? What are the attributes of that data that drive your model? What ML platforms are you using?” Those are excellent questions — the answers to which build a good business case for leveraging machine learning.
What is machine learning?
Ronald Schmelzer (@rschmelzer), Managing Partner and Principal Analyst at Cognilytica, notes, “AI isn’t a discrete technology. Rather it’s a series of technologies, concepts, and approaches all aligning towards the quest for the intelligent machine.”[2] One of those technologies is machine learning. Sophie Hand, UK country manager at EU Automation, explains, “Machine learning is a subset of artificial intelligence where computers independently learn to do something they were not explicitly programmed to do. They do this by learning from experience — leveraging algorithms and discovering patterns and insights from data. This means machines don’t need to be programmed to perform exact tasks on a repetitive basis.”[3] Some subject matter experts see red when they hear or read people equating machine learning to artificial intelligence. Why? Alan J. Porter (@alanjporter), Head of Strategic Services for [A], explains scientists understand “ML is great at recognizing patterns but not much else” and “ML assumes tomorrow is going to be the same as today.”[4]
In a previous article, Schmelzer writes, “For the layperson, we want to stress that AI is not interchangeable for ML and certainly ML is not interchangeable with Deep Learning. But ML supports the goals of AI, and Deep Learning is one way to do certain aspects of ML. Or to put it another way, doing machine learning is necessary, but not sufficient, to achieve the goals of AI, and Deep Learning is an approach to doing ML that may not be sufficient for all ML needs.”[5] In his subsequent article, Schmelzer explains recognizing patterns is not the same as understanding what’s being analyzed. He writes, “Without understanding, there’s no common sense. Without common sense and understanding, machine learning is just a bunch of learned patterns that can’t adapt to the constantly evolving changes of the real world.”
Towards machine reasoning
Schmelzer believes one of the next steps towards artificial intelligence is developing machine reasoning. Machine reasoning, which involves understanding and common sense, requires an ontology. He notes, “In 1984, the world’s longest-lived AI project started. The Cyc project is focused on generating a comprehensive ‘ontology’ and knowledge base of common sense, basic concepts and ‘rules of thumb’ about how the world works. The Cyc ontology uses a knowledge graph to structure how different concepts are related to each other, and an inference engine that allows systems to reason about facts. The main idea behind Cyc and other understanding-building knowledge encodings is the realization that systems can’t be truly intelligent if they don’t understand what the underlying things they are recognizing or classifying are.” I couldn’t agree more. For that reason, Enterra Solutions® partnered with Cycorp® early in our development of the Enterra Cognitive Core™ — a system that can Sense, Think, Act and Learn®. As Schmelzer notes, “We need more than machine learning — we need machine reasoning. … Indeed, we’re rapidly facing the reality that we’re going to soon hit the wall on the current edge of capabilities with machine learning-focused AI. To get to that next level we need to break through this wall and shift from machine learning-centric AI to machine reasoning-centric AI.”
As machine reasoning matures, Steve Phillpott (@steve_phillpott), CIO at Western Digital, predicts, “Virtually anything and everything that can be automated, will be. New AI/ML models and insights, leveraging the convergence of multiple data types, will be the key enablers of automation.”[6] Like Glode, Nicole Martin, owner of NR Digital Consulting, asserts it’s important to understand the limits of machine learning and how it differs from artificial intelligence. “The important thing to remember with ML,” she writes, “is that it can only output what is input based on the large sets of data it is given. It can only check from what knowledge it has been ‘taught.’ If that information is not available, it cannot create an outcome on its own.”[7] On the other hand, she explains, “AI can create outcomes on its own and do things that only a human could do. ML is a part of what helps AI by taking the data that it has been learned and then the AI takes that information along with past experiences and changes behavior accordingly.” She adds, “They are both crucial to the future of technology.” For business leaders who want to make the most of cognitive technologies, understanding the differences between AI and machine learning does matter.
Concluding thoughts
When considering what kind of cognitive technology your company needs, you start with Glode’s first question: “What task are you addressing or automating, and what problem are you solving?” If you are trying to memorialize and learn from the past, machine learning probably works. Kevin Gardner, a business consultant for InnovateBTS, explains, “The primary role and aim of ML are to improve efficiency through experience. ML is focused on ensuring that a machine that performs a particular task can perform it correctly without concern for success.”[8] If you want to improve decision-making, then you need to leverage machine reasoning. Cognitive technologies are often described as decision support systems. Gardner writes, “The primary aim of AI is to increase the probability of success and not accuracy.” If that sounds counterintuitive, you have to understand the business landscape is colored in many shades of gray, not just black and white. Decisions must be made in this complex and ambiguous environment and cognitive technologies (i.e., reasoning machines) can help decision-makers improve chances of making more successful decisions.
Footnotes
[1] Christopher Glode, “Machine Learning And Your Company: Hype Or Hope?,” Forbes, 19 November 2019.
[2] Ronald Schmelzer, “Going Beyond Machine Learning To Machine Reasoning,” Forbes, 9 January 2020.
[3] Sophie Hand, “What is the role of machine learning in industry?” Engineer Live, 14 January 2020.
[4] Alan J. Porter, “Machine Learning Isn’t Rocket Science,” CMS Wire, 9 September 2019.
[5] Ronald Schmelzer, “Is Machine Learning Really AI?” Forbes, 21 November 2019.
[6] Perry Cohen, “Expert Predictions for 2020 Part 4: AI and Machine Learning,” Embedded Computer Design, 27 December 2019.
[7] Nicole Martin, “Machine Learning And AI Are Not The Same: Here’s The Difference,” Forbes, 19 March 2019.
[8] Kevin Gardner, “Clearing the Confusion: Artificial Intelligence and Machine Learning,” BBN Times, 6 September 2019.
