Machine learning is a term that has various synonyms - names that are the result of either marketing activities by corporates or just terms that have been used interchangeably. Although some may argue that they have different implications, they all ultimately refer to machine learning as a subject that facilitates the prediction of future events using historical information.

The commonly heard terms for machine learning include predictive analysis, predictive analytics, predictive modeling, and many others. As such, unless the entity that publishes material explaining their interpretation of the term and more specifically, how it is different, it is generally safe to assume that they are referring to machine learning. This is often a source of confusion among those new to the subject, largely due to the misuse and overuse of technical verbiage.

Statistics, on the other hand, is a distinct subject area that has been well known for over 200 years. The word is derived from the new Latin, statisticum collegium (council of state, in English) and the Italian word statista, meaning statesman or politician. You can visit https://en.wikipedia.org/wiki/History_of_statistics#Etymology for more details on this topic. Machine learning implements various statistical models, which due to the rigor of computation involved, is distinct from the branch of classical statistics.

AI is also closely related to machine learning, but is a much broader subject. It can be loosely defined as systems (software/hardware) that, in the presence of uncertainties, can arrive at a concrete decision in (usually) a responsible and socially aware manner to attain a target end objective. In other words, AI aims to produce actions by systematically processing a situation that involves both known and unknown (latent) factors.

AI conjures up images of smart and sometimes rebellious robots in sci-fi movies, just as much as it reminds us of intelligent systems, such as IBM Watson, that can parse complex questions and process ambiguous statements to find concrete answers.

Machine learning shares some of the same traits - the step-wise development of a model using training data, and measuring accuracy using test data. However, AI has existed for many decades and has been a familiar household term. Institutions in the US, such as Carnegie Mellon University, have led the way in establishing key principles and guidelines of AI.

The online resources/articles on AI versus machine learning do not seem to provide any conclusive answers on how they differ. However, the syllabus of AI courses at universities makes the differences very obvious. You can learn more about AI at https://cs.brown.edu/courses/csci1410/lectures.html.

AI refers to a vast array of study areas that involve:

• Constrained optimization: Reach best possible results given a set of constraints or limitations in a given situation
• Game theory: For instance, zero-sum games, equilibrium, and others - taking a measured decision based on how the decision can affect future decisions and impact desired end goals
• Uncertainty/Bayes' rule: Given prior information, what is the likelihood of this happening given something else has already happened
• Planning: Formulating a plan of action = a set of paths (graph) to tackle a situation/reach an end goal
• Machine learning: The implementation (realization) of the preceding goals by using algorithms that are designed to handle uncertainties and imitate human reasoning. The machine learning algorithms generally used for AI include:
  • Neural networks/deep learning (find hidden factors)
  • Natural language processing (NLP) (understand context using tenor, linguistics, and such)
  • Visual object recognition
  • Probabilistic models (for example, Bayes' classifiers)
  • Markov decision processes (decisions for random events, for example, gambling)
  • Various other machine learning Algorithms (clustering, SVM)
• Sociology: A study of how machine learning decisions affect society and take remedial steps to correct issues