Artificial Intelligence

1. Explain how Intelligent Agents influence the development of applications or Software.

The Intelligent Agents are used in the creation of machines (hardware / software) to display intelligent behavior. Such systems can be simply personal computers; or may be robots. To perform tasks that if a human's conduct, require hard work and a high level of concentration and intelligence. 2. Identify and explain as a model of software engineering for distributed systems using Intelligent Agents.

The Software Engineering model uses the following features of an Intelligent Agent: reasoning abilities, skills, services, sensors, effectors, groups they belong and class. ABLE is an example; is a tool for building IBM Systems Intelligent Agents where its entirety, including agents, are constructed by composition AbleBeans, an extension of JavaBeans. Interest are a set of specialized systems that implement AbleBeans statistical learning (self-organizing maps, neural networks, connection maps) and symbolic control (forward chaining reasoning and predicate logic). ABLE also incorporates a visual development environment where AbleBeans interconnect. The interest of this platform is that visually solves the construction and communication of agents. 3. When does an agent is considered autonomous?

The main point about the agent is that they are independent: able to act independently, exhibiting control over their internal state.

4. Analyze each of the following techniques to implement rationality:

Knowledge-Representation Deals with how to shape the operational knowledge of how computer with sufficient intelligibility (that can be understood) and flexibility. It is a description in any language part of the real world, its condition and the knowledge we have about him. Knowledge is twofold: first, facts and other rules.

Symbolic reasoning represents knowledge through symbols and relations between them. Specific algorithms that process the symbols to solve problems or infer new knowledge is used.

Search process planning- and articulation of a sequence of actions to reach a goal.

Satisfaction restrictions- Find a state that satisfies the constraints (assigning values to the variables that satisfies the constraints).

Learning Automatic- Agents / learning programs or evolve based on their experience, to perform a particular task getting better. The main goal of any learning process is to use the evidence known to create a hypothesis and able to respond to new situations unknown.

Then explain how each contributes to the development of rationality within the term of Intelligent Agents.

The Agent always done "right" data from perceiving the environment. That is why rationality in each one helps each intelligent agent may have the ability to think logically. 5. Explain the difference between Communication Agents and Distribution Agents.

Communication Agents are available in almost all Multi-Agent Systems. Provide support to users through user interaction. They help the user in performing a task in an application and carry out the instructions of the user. In situations that do not have exact instructions to follow these agents act in accordance with established knowledge through user feedback; They can also seek help from other agents.

The main features Distribution Agents are adaptability and distribution. They use intelligent software agents to provide the key to dynamically adapt content to the needs of people intelligence.

6. Explain how each affects the automation of processes.

Process automation not only increases the time we can devote to other tasks, but it eliminates many manual processes that can lead to errors. Communication Agents provide useful assistance to humans. For example, email filters learn to choose e-mails rubbish indications that the user experience. Meanwhile, the Distribution Agents can be used to improve the market of finance, such as through an online store.

7. Explain the term perceptual sequence Ideal Rational Agent.

All that the agent has received so far. What the agent knows about the environment. The actions that the agent can perform.

8. Explain the conceptual properties intone an Agent based on the following: Action accessible-is one in which the agent can obtain exact-date on the status of the environment, complete information.

Inaccessible Action- Most environments of moderate complexity (including, for example: the physical world every day and the Internet) are inaccessible.

Deterministic action is one in which every action has a single guaranteed effect - no uncertainty about the state resulting from an operation.

Indeterministic Action- They have bigger problems for the designer Agent.

Episodic Agent- In an episodic environment, the performance of an agent depends on a number of discrete episodes, with no link between performance of an agent in different scenarios.

Episodic environments are simpler from the perspective of the developer agent because the agent can decide what action to take, based only on the current episode - no need to reason about the interactions between this and future episodes.

Not Episodic Agent- No experience can be divided into atomic Agent episodes.

Static Agent- can remain unchanged, except for the implementation of actions by the Agent.

Dynamic-agent has other processes that operate on it, and therefore changes in ways beyond the control of the agent.

A discreet-agent environment is discrete if there is a fixed finite number of actions and perceptions in it.

Agent continuous sequential Perceptions continued.

9. Describe and discuss the function of a based-on-Goals agent.

An Agent Based-on-Goals is the description of a situation we want to achieve a set of properties that we want to keep (when the agent succeeds in its goal). For this it is necessary to define a test shot that captures what it means to have reached the goal.
It's easy when with one goal is achieved with an action. Must be flexible with respect to target different destinations as dialing a new destination, the agent creates a new behavior. Goals can present a future desired state or situation, value or a set of rules or restrictions. Therefore, a powerful data mapping capability built into the engine and process and requires no external rules.