Difference between Artificial Intelligence and Machine Learning

“Artificial Intelligence” and “Machine Learning” these two terms have become ubiquitous these days. They are often used interchangeably, however, technically speaking, these two terms happen to be different.

Artificial Intelligence (AI) is a branch of computer science that focuses on making machines learn and perform intelligent tasks like humans.

AI doesn’t necessarily mean robots performing tasks for us. Rather, it is already part of our daily lives in the form of algorithms and models used in the back-end by various products we use.

AI includes algorithms that can be trained to generate required output from given input

Whenever someone uses google translator to translate text from one language to another or chooses the optimal path suggested by Google maps, that person is interacting with an AI system.

Machine Learning (ML) is the set of techniques machines use to learn from the data without being specifically programmed by domain expertise. ML is a subset of AI that helps it achieve its vision of intelligent machines.

ML finds a range of use cases today and its usage is expected to increase manifolds in the future.

Google and Yahoo use ML for email spam and malware filtering. Their models don’t merely consist of a set of rules to classify an email as spam, but they generate new rules themselves based on the new data they deal with. Similarly, whenever someone watches a movie on Netflix recommended by it, that person is interacting with an ML-based AI system.

ML and AI application

ML and AI application

ML and AI application

Beyond ML, AI has many other domains as well. Reinforcement Learning (RL) is one such domain that doesn’t require a system to learn the underlying concepts before making a decision. It focuses on a carrot and stick approach, where the algorithm incentives a specific type of response while punishing an unfavourable response. Natural Language Processing (NLP) is another domain of AI that focuses on developing language models for various use cases. Text summarization, converting voice to text, and language translation are some of the use cases of NLP. NLP includes both ML and statistics-based (non-ML) based models.

All major industries today are utilising AI and ML to achieve operational efficiencies. Industry-wise key use cases of AI include:

1. Health Care Industry: AI and ML is being used in healthcare for predictive medication. Based on the patient’s medical history, these technologies are predicting the likelihood of a patient developing a serious disease much before any actual symptom is shown. This has the potential of saving millions of lives across the world.
2. Banks and Financial Institution: AI and ML is helping bankers develop models to better judge the creditworthiness and default risk of an individual. This can significantly improve the profitability of banks.
3. Defence: AI and ML are helping researchers develop autonomous defence systems. Machines can take over the roles of humans in sensitive operations.

We are heading to a future surrounded by intelligent machines. Only time can tell what possibilities this will unravel for us.