人工智能解决的问题_人工智能中的问题解决

人工智能解决的问题

The aim of Artificial Intelligence is to develop a system which can solve the various problems on its own. But the challenge is, to understand a problem, a system must predict and convert the problem in its understandable form. That is, when an agent confronts a problem, it should first sense the problem, and this information that the agent gets through the sensing should be converted into machine-understandable form. For this, a particular sequence should be followed by the agent in which a particular format for the representation of agent's knowledge is defined and each time a problem arises, the agent can follow that particular approach to find a solution to it.

人工智能的目的是开发一种可以自行解决各种问题的系统。 但是挑战在于，要理解问题，系统必须以可理解的形式预测并转换问题。 就是说，当一个代理遇到问题时，它应该首先感知该问题，并且该代理通过感知获得的信息应转换为机器可理解的形式。 为此，代理应该遵循特定的顺序，在该顺序中定义了用于表示代理知识的特定格式，并且每次出现问题时，代理都可以遵循该特定方法来找到解决方案 。

The steps involved in solving a problem (by an agent based on Artificial Intelligence) are:

解决问题 (通过基于人工智能的代理 )所涉及的步骤是：

1) Define a problem

1)定义一个问题

Whenever a problem arises, the agent must first define a problem to an extent so that a particular state space can be represented through it. Analyzing and defining the problem is a very important step because if the problem is understood something which is different than the actual problem, then the whole problem-solving process by the agent is of no use.

每当出现问题时，代理必须首先将问题定义到一定程度，以便可以通过它表示特定的状态空间。 分析和定义问题是非常重要的一步，因为如果理解问题与实际问题有所不同，那么座席解决问题的整个过程将毫无用处。

2) Form the state space

2)形成状态空间

Convert the problem statement into state space. A state space is the collection of all the possible valid states that an agent can reside in. But here, all the possible states are chosen which can exist according to the current problem. The rest are ignored while dealing with this particular problem.

将问题陈述转换为状态空间。 状态空间是代理可以驻留的所有可能有效状态的集合。但是在这里，根据当前问题选择可以存在的所有可能状态。 在处理此特定问题时，将忽略其余部分。

3) Gather knowledge

3)收集知识

collect and isolate the knowledge which is required by the agent to solve the current problem. This knowledge gathering is done from both the pre-embedded knowledge in the system and the knowledge it has gathered through the past experiences in solving the same type of problem earlier.

收集并隔离代理解决当前问题所需的知识。 知识的收集既可以从系统中预先嵌入的知识中进行，也可以通过它在解决早期类型问题时的过去经验中收集而来。

4) Planning-(Decide data structure and control strategy)

4)规划-(确定数据结构和控制策略)

A problem may not always be an isolated problem. It may contain various related problems as well or some related areas where the decision made with respect to the current problem can affect those areas. So, a well-suited data structure and a relevant control strategy must be decided before attempting to solve the problem.

问题不一定总是孤立的问题。 它也可能包含各种相关问题，或者某些与当前问题有关的决策可能影响这些领域的相关领域。 因此，在尝试解决该问题之前，必须确定合适的数据结构和相关的控制策略。

5) Applying and executing

5)应用和执行

After all the gathering of knowledge and planning the strategies, the knowledge should be applied and the plans should be executed in a systematic way so s to reach the goal state in the most efficient and fruitful manner.

在所有知识的收集和战略计划之后，应该应用知识并以系统的方式执行计划，从而以最有效和富有成果的方式达到目标状态。

翻译自: https://www.includehelp.com/ml-ai/problem-solving-in-artificial-intelligence.aspx

人工智能解决的问题