Trending Technology Machine Learning, Artificial Intelligent, Block Chain, IoT, DevOps, Data Science

Recent Post

Search

Monday, 18 June 2018

Environments in Artificial Intelligence

Environments
Actions are done by the agent on the environment.
Environment provides precepts to the agent.
Determine to a large degree the interaction between the "outside world" and the agent
   - the "outside world" is not necessarily the "real world" as we perceive it.
It may be a real or virtual environment the agent lives in
In many cases, environment are implemented within computers
   - They may or may not have a close correspondence to the "real world".

Properties of environments

 Full observable vs. partially observable
Or Accessible vs. inaccessible
    - If an agent's sensory equipment gives it access to the complete state of the environments, then we say that environment is fully observable to the agent.
    - An environment is effectively fully observable if the sensors detect all aspects that are relevant to the choice of action.
    - A fully observable environment is convenient because the agent need not maintain any internal state to keep track of the world.

Deterministic vs. nondeterministic.
   - If the next state of the environment is completely determined by the current state and the actions selected by the agents, then we say the environment is deterministic.
   - If the environment is inaccessible, then it may appear to be nondeterministic (bunch of uncertainties).

Properties of task environments

Episodic vs. sequential
   - Agent's experience is divided into "episodes".
       Each episode consists of the agent perceiving and acting.
   - Subsequent episodes do not depend on what actions occur in previous episodes.
   - In sequential environments current actions affect all succeeding actions.

Static vs. Dynamic
   - If the environment can change while an agent is performing action, then we say the environment is dynamic.
   - Else its static.
   - Static environments are easy to deal with, because the agent does not keep on looking at the environment while it is deciding on an action.
   - Semidynamic :  if the environment does not change with the passage of time but the agent performance score does.


Single agent vs. Multiagent
   - In the single agent environment there is only one agent
          A computer software playing crossword puzzle
   - In multiagent systems, there are more than one active agents
          video game



 

No comments:

Post a Comment