We were reading TechRepublic's recent article explaining the differences between artificial intelligence, machine learning and deep learning and thought you'd enjoy it. Here are our thoughts.
TechRepublic positions artificial intelligence as a technology that will revolutionize business, then acknowledges it is frequently used interchangeably with machine learning and deep learning. They want to educate readers about the three technologies and coin the term "machine intelligence" as what readers will master by the end. We personally love this term.
Artificial intelligence (AI)
They describe artificial intelligence as broad: "AI can refer to anything from a computer program playing a game of chess, to a voice-recognition system like Amazon's Alexa interpreting and responding to speech." AI also has three categories that differ by the number of skilled tasks performed and intelligence level.
This is an accurate portrayal of AI, but depending on who you talk to, AI is categorized differently. We recently attended a talk at SaaStr Annual where Greylock Partners discussed four business models for AI-enabled software. One is replacing rules-based systems, which is exactly what our machine learning platform does.
Machine learning (ML)
Machine learning is a branch of artificial intelligence. TechRepublic proclaims ML is "the most promising tool in the AI kit for businesses." ML enables systems to recognize patterns in data which can cure the aches and pains of "big data."
The article links to another one about how enterprise companies can implement ML, but ML is also extremely useful for small and medium-sized business (SMB). Our mission is to democratize machine learning by making it accessible to retailers big and small.
Deep learning is a subset of ML. TechRepublic says "Deep learning uses some ML techniques to solve real-world problems by tapping into neural networks that simulate human decision-making." It is also expensive and requires massive data sets to learn.
Overall they explain deep learning well, especially since they point out it is suspectible to bias by giving a real-world example from Google.