Provided by Know Labs in Partnership with Stanford University’s Engineering Department
“Online Introduction to Artificial Intelligence is based on Stanford CS221, Introduction to Artificial Intelligence. This class introduces students to the basics of Artificial Intelligence, which includes machine learning, probabilistic reasoning, robotics, and natural language processing.
The objective of this class is to teach you modern AI. You learn about the basic techniques and tricks of the trade, at the same level we teach our Stanford students. We also aspire to excite you about the field of AI. Whether you are a seasoned professional, a college student, or a curious high school student – everyone can participate.
This online class will make this material available to a worldwide audience. But rather than just watching lectures online, you will participate. You will do homework assignments, take exams, participate in discussions with other students, ask questions of the instructors, and also get a final score.”
Source – ai-class.com
- An intelligent agent interacts with an environment.
- It detects the state of its environment through its sensors.
- It can effect the state of the environment through its actuators.
- The function that recieves its input (sensors) and responds with an output (actuators) is known as the control policy for the agent.
- The control policy creates a loop in which:
Input data is recieved (sensors)
Input data is processed, a decision is made how to respond to this data. (Control Policy)
Actuators send data to environment in an effort to change the environment.
- The previous loop is known as the Perception Action Cycle
From the earliest stages of childhood development, we are fascinated with both ourselves and our environment. It would seem that we learn as a result of discovering who we are rather than our environments response to what we do.
Considering this, I propose the following concept:
An intelligent agent may have greater success not through the awareness of it’s environment; but rather, an intelligent agent should see all factors as individual pieces of its environment, including itself. Each factor in its environment has a corresponding equation, with all equations culminating together into a much greater equation.
Artificial Intelligence Terminology
- Fully versus Partially (observable)
An environment is considered fully observable if the intelligent agent can always see (input data) the full state of the environment. An environment is partially observable if the agent can only see a portion of the environment, yet it is able to calculate past environments to better understand its current state. (i.e. Playing Blackjack 21 using 52 cards. The agent can “count cards” and knows which cards have not yet been played)
- Determinist versus Stochastic
A deterministic environment is one where an agent’s actions uniquely determine the outcome. (For example, in the game of chess, the agent move piece to a set place.) A Stochastic environment is one where the agents actions to not directly determine the outcome. (For example, in a game of dice, the roll is random and the agent cannot determine the outcome.)
- Discrete versus Continuous
A discrete environment in one in which there are a finite number of actions to be performed. (Chess for example). A continuous environment is one in which the number of actions that can be performed maybe infinite. (For example, to play a game of darts, there are an infinte number of angles and infinite number of accelerations.)
- Benign versus Adversarial
Benign environments are ones that may be random or stochastic, yet the environment has not objective that may contradict the agents objective. (Weather is random and it may affect your reaction, but it’s purpose is not to negatively affect someone.) An adversial environment is one where the environment is “out to get” the agent. (Chess is an example).
Artificial Intelligence and Uncertainty Management
What to do when you do not know what to do?
Reasons for Uncertainty:
- Sensor Limits
- Stochastic Environments
Intro to Artificial Intelligence – Summary