Introduction to Artificial Intelligence

State the 3 Laws of Robotics

A robot may not injure humans A robot must obey A robot must protect its existence

What does it mean when the word "smart" is in the name of a product.

AI is applied.

Weak AI (Android, iRobot, Atlas of Boston Dynamics)

Act like people

Considered as strong AI

Act rationally

Enables machine to think like people and make it intelligent

Artificial Intelligence

It deals with simulation of intelligent behavior and mimics cognitive functions of human

Artificial Intelligence

Column to be predicted

Target

How computers can be made to identify objects, scenes, and activities from images, including image processing, recognition, detection, analysis, and understanding.

Computer Vision

It is the computation of multilayer neural networks feasible.

Deep Learning

What is the relationship of AI, ML, and DL

Deep Learning (DL) is subset of Machine Learning (ML) is subset of Artificial Intelligence (AI)

What are the examples of Machine Learning

Emails (spam or not)

Is AI at its final stage.

FALSE, AI is still in its initial age.

These are attributes of the data to come into an outcome, used for predictions

Features

What is Artificial Intelligence

Technique that enables computers to mimic human activities

It learn patterns as it is exposed to more data over time.

Machine Learning

What do you call chatbots, task-driven dialogs, search engine, machine translation, etc.

Natural language processing

Two main types of ML and its definition.

Supervised Learning = has target column ; make predictions; Example is Fraud Detection Unsupervised = no target column; finds structure in the data. Example is customer segmentation

Considered as strong AI but unavailable yet due to limitations in brain science.

Think rationally.

Type of AI considered as weak. Examples of it are Watson and AlphaGo

Think like people

What are the four types of AI

Think like people Act like people Think rationally Act rationally (TATA 2 people, 2 rationally)

Deep Learning combines steps in Machine Learning into a complex neural network.

True

ML is limited to identifying a target with same color.

True