Artificial Intelligence (AI)

Where AI is today, it can solve very specific tasks such as

• Playing Chess
• Counting the number of objects on a conveyer belt

The AI has learned concepts it has learned across multiple domains. This is what people do and AI can not do this today!!

Going beyond Artificial General Intelligence where the machines learn to communicate with themself and actively drive change independently of humans.

From our respective, the Artificial Super intelligence would be equivalent to a god.

Deep learning uses something that is called a neural network. The original design of a neural network took inspiration from the connections in the human mind to create a mathematical formulation with a large number of parameters that can be learned to do a specific task.

Machine learning uses techniques from Deep Learning, statistics, and other computer algorithms to imitate how humans learn to perform a specific task.

When applied to a specific task, such as image classification where one attempts to teach the computer how to label what is in an image. t can learn how to classify different dog breeds at a superhuman level. But using the same model that is specialized on dog breeds, to predict if it is a cat, dog, or a horse, it would fail spectacularly unless it got retrained for the new task.

Machine Learning enables a computer to do a specific task such as detecting if something is in an image. The next level is to use the results and perform decision-making and let it take direct action.

For example, it could be something as simple as on a production line, detecting if there is a defect in what is being produced and discarding the damaged item, this is referred to as defect detection.

The computer learns to understand speech. There are several different subtasks such as directly understanding speech such as Apples Siri known as automatic speech recognition (ASR). Another common task is to simply convert the speech to text.

An online retailer needs to make relevant product recommendations, here is where Artificial intelligence comes in. Using historical patterns in the user consumption data the AI can come up with the ideal recommendations for effective cross-selling.

The ability to help customers as fast as possible at any time of the day is important for customer satisfaction. That is why chatbots are replacing humans more and more. They can easily answer frequently asked questions such as shipping rates, shipping times, suggested sizes, etc.