You are on page 1of 4

COURSE TITLE: INTRODUCTION TO ARTIFICIAL INTELLIGENCE

MODULE TITLE: ADDING NEW HUMAN OCCUPATIONS

Chapter 17 – ADDING NEW HUMAN OCCUPATIONS

One of the most exciting aspects of entering a new field of human endeavor is the
learning curve that is created by the fact that we have not done the things we are
doing now. You might be able to make a significant contribution to humanity if you
are willing to face the challenge of discovering and taking risks associated with
doing something different.
Understanding an algorithm's meaning An algorithm will always present a series of
steps for solving a problem, though it may not complete all of these steps.
Algorithms cover a variety of topics. Data can be stored, analyzed, and arranged in
data structures in a variety of ways. Algorithms are capable of resolving issues in
communication, industrial production and supply, finance, science, and medicine.

Every algorithm is a series of steps taken to quickly solve a problem (or report back
if there is no solution). Problems solved by AI algorithms are thought to typically
(or even exclusively) result from human intelligent behavior, which sets them apart
from generic algorithms. The class of problems known as NP-complete, where NP
denotes nondeterministic polynomial time, is one that AI algorithms frequently
encounter. Most of the time, people approach these problems by combining their
sense of reason and intuition. A few examples include:

Recognizing patterns in image vision (in contrast to something like image


restoration or image processing) » Recognizing patterns in image vision (in
contrast to something like image restoration or image processing) » Rec Even if you
had computers that were more powerful than those that are currently available,
COURSE TITLE: INTRODUCTION TO ARTIFICIAL INTELLIGENCE
MODULE TITLE: ADDING NEW HUMAN OCCUPATIONS

the search for the answer would take nearly indefinitely. This kind of issue is
referred to as AI-complete in artificial intelligence.

Algorithms and data structures that are unique to working with a state space are
required. Trees and graphs are the fundamental data structures that are used the
most. The breadth-first search or the deep-first search algorithms are the most
frequently used ones for effectively exploring graphs.

A graph is a type of structure that is composed of a number of nodes—also known


as vertexes—connected by a number of edges or arcs, depending on how they are
displayed. Consider a structure that is similar to a map when thinking about a
graph: The streets represent the edges, and a node is a location on the map. In
contrast to a tree, where each path reaches a leaf node, this presentation is distinct.
For an illustration of a graph, see Figure 3-2. When determining states that are
analogous to physical spaces, graphs are especially helpful. The GPS, for instance,
uses a graph to show streets and locations.

It's possible that you were unaware of some of the novel twists that graphs
introduce. A graph, for instance, might contain the idea of directionality. A graph
node can connect to any other node in a particular direction, in contrast to a tree,
which has parent-child relationships. Take a look around a city's streets. While
most streets allow traffic to go either way, some only allow one way traffic.

It's possible that the actual connections on the graph don't show exactly how they
are presented. The weight of a connection can be calculated by using a graph. The
weight could tell you anything about the distance between the two points, how long
it will take to travel the route, or anything else.
COURSE TITLE: INTRODUCTION TO ARTIFICIAL INTELLIGENCE
MODULE TITLE: ADDING NEW HUMAN OCCUPATIONS

A tree is nothing more than a graph with only one path connecting any two vertices
and no cycles (the ability to return to the parent from any child). Numerous graph
algorithms are only applicable to trees.

vertex is either discovered (because you just visited it) or processed (because the
algorithm tried all of its edges). The type of search that is carried out is determined
by the order of the search: informed (heuristic) and uninformed (blind search). The
AI's sole source of information in the uninformed strategy is the graph structure it
discovers while traversing the state space. Two common blind search algorithms
are discussed in the following sections: A breadth-first search, or BFS, examines
every node that connects to the graph's root. These searches put breadth ahead of
depth. After that, it moves on to the next level and explores each one separately
until it reaches the end. BFS looks at the graph in a systematic way, rotating around
the starting vertex at each vertex. As a result, the search examines A, B, and C first
in the example graph, then D. It visits all vertices in a single step, starting at the
starting vertex; After that, it moves on to advance to 2, 3, and so on.

From the graph's root, a depth-first search, or DFS, follows a single path to each
node. It then retraces its steps and begins investigating additional options until it
reaches the root. The algorithm then reengages in the same search process and
selects one of the available root alternatives. The objective is to thoroughly
investigate each path before proceeding to the next one.

State-space search adversarial games are interesting because they show AI's current
capabilities and opportunities for the future. This is the scenario in adversarial
games, which are games in which one player wins and the others lose, as well as
similar situations in which players aim for opposing goals. A great illustration of a
space search game is a straightforward game like tic-tac-toe, which you may have
already seen an artificial intelligence play. In the 1983 film WarGames, the
supercomputer WOPR (War Operation Plan Response) plays against itself at a
COURSE TITLE: INTRODUCTION TO ARTIFICIAL INTELLIGENCE
MODULE TITLE: ADDING NEW HUMAN OCCUPATIONS

breakneck pace, but it can't win because a state-space search can never lose.

Utilizing heuristics and local search The state-space search strategy has many
moving parts. No machine, no matter how powerful, can ultimately list all of a
situation's possibilities. This section continues with games because they are
predictable and have clear rules, whereas many real-world situations are
unpredictable and lack clear rules. As a result, games make people feel good.

Local search solutions may not find a solution in a reasonable amount of time (you
can stop the process and use the current solution) or produce a solution of poor
quality, despite their straightforward design. Even though there is no guarantee
that a local search will find a solution to a problem right away, you can increase
your chances of doing so if you give it enough time to complete its calculations.
When it can't improve the solution in any other way, it stops. The most crucial
factor is selecting the appropriate location for exploration.

You might also like