Programme Name: ________BCs.(Hons.

)_______________________

Course Code: ___CSC 3201_______

Course Name: __________Artificial Intelligence_____________

Assignment / Lab Sheet / Project / Case Study No: _1__

Date of Submission: _______2021/07/13______________________

Submitted By: Submitted To:

Student Name: Reshma Oli Faculty Name: Prakash Chandra

IUKL ID: 041902900054 Department: P.O.

Semester: 4th

Intake: September
1. Explain the concept of artificial intelligence. Explain the application of AI in various
fields.

Ans:-

Artificial intelligence (AI), also known as machine intelligence, is a branch of computer

science that aims to imbue software with the ability to analyze its environment using
either predetermined rules and search algorithms, or pattern recognizing machine
learning models, and then make decisions based on those analyses.

While AI often invokes images of the sentient computer overlord of science fiction, the
current reality is far different. At its heart, AI uses the same basic algorithmic functions
that drive traditional software, but applies them in a different way.

Artificial intelligence can be allowed to replace a whole system, making all decisions
end-to-end, or it can be used to enhance a specific process.

For example, analyzing video footage to recognize gestures, or replacing peripheral

devices (keyboard, mouse, touchscreen) with a speech to text system., giving the
impression that one is interacting with a sentient being.

Just as philosophers debate the nature of man and the existence of free will, computer
science experts debate the various types of AI.

Application of AI

AI is currently being applied to a range of functions both in the lab and in

commercial/consumer settings:

A. Speech Recognition

Allows intelligent systems to convert human speech into text or code. Speech
recognition can be divided into three aspects:

● Speech synthesis, including online and offline speech synthesis;

● Speech recognition, including speech dictation and other aspects;
● Semantic understanding is to use neural networks to extract the meaning
of speech, including voice evaluation and some features of some of our
commonly used machine translation.
B. Natural Language Processing

A subset of speech recognition, enables conversational interaction between

humans and computers.
 ● Computer translation: It mainly includes machine simultaneous
translation.
● Information Retrieval: For example, when I tell the computer what I am
looking for, then it can search related things for me.
● Intelligent customer service: We interact with the computer through
voice and let it answer our questions.

C. Computer Vision

Allows a machine to scan an image and identify it using comparative analysis.

Perhaps the most revolutionary aspect of AI, however, is that it allows software to
rewrite itself as it adapts to its environment.

Unlike traditional upgrade programs that take years and are often buggy, or even
newer DevOps processes that push changes quickly with less disruption, AI
allows a given program to optimize itself to highly specialized use cases.

This should not only lower the cost of software licensing and support, it should
provide steadily improving performance and the development of unique processes
that deliver crucial advantages in an increasingly competitive economy.

D. Health Care Industry

The applications of artificial intelligence can be seen in a wide variety of

applications ranging from preparing prescriptions until conducting complex
operations. A group of MIT researchers has built a tiny robot that unfolds itself in
the stomach and treats swallowed button batteries, wounds. Artificial Intelligence
is also used to prepare a prescription for patients based on individual medical
conditions. Besides, Artificial Intelligence is also used in assisting the physicians
to diagnose reports accurately.

E. Gaming Industries

The quality of a game cannot be measured alone by the user interface. Rather, a
good game requires a good Artificial intelligence inbuilt for the characters. This is
more accurate to any combat games like Battlefield, and Destiny. These games
top the list because of their complex Artificial Intelligence algorithm. Gaming
companies like RockStar and Ubisoft deliver games with great Artificial
Intelligence algorithms. Thus, these companies make a good market compared to
their competitors.
2. Explain the concept of intelligence agent? With examples explain different types of
intelligent agents.

Ans:-

Intelligent Agents can be any entity or object like human beings, software, machines.
These agents can make decisions based on the inputs it receives from the environment
using its sensors and act on the environment using actuators. AI-Enabled agents collect
input from the environment by using sensors like cameras, microphone or other sensing
devices. The agents perform some real-time computation on the input and deliver output
using actuators like screen or speaker. These agents have abilities like Real-Time problem
solving, Error or Success rate analysis and
information retrieval.

Intelligent Agent can come in any of the three

forms, such as:-

● Human-Agent
● Robotic Agent
● Software Agent

Types:-

1. Simple Reflex Agents

They are the basic form of agents and function

only in the current state. They have very low
intelligence capability as they don’t have the
ability to store past state. These types of agents respond to events based on predefined
rules, which are pre-programmed. They perform well only when the environment is fully
observable. These agents are helpful only in a limited number of cases, something like a
smart thermostat. Simple Reflex Agents hold a static table from where they fetch all the
predefined rules for acting.

For example

An industrial robot working on part transfer across machines is a good example. An

action or series of actions are triggered when a proximity sensor detects its presence at a
spatial node. They usually do not store their past percepts and decide on the current
actions.

Another such example is iDraw, a drawing robot. It directly converts the typed in
characters into writing, without requiring to store past percept information.
2. Model-Based Agents

It is an advanced version of the Simple Reflex agent. Like Simple Reflex Agents, it can
also respond to events based on the predefined conditions; on top of that, it can store the
internal state (past information) based on previous events. Model-Based Agents update
the internal state at each step. These internal states aid agents in handling the partially
observable environment. To perform any action, it relies on both internal state and current
percept. However, it is almost impossible to find the exact state when dealing with a
partially observable environment.

For example

An automated taxi may not be able to see around the large truck that has stopped in front
of it and can only guess about what may be causing the hold-up. Thus, uncertainty about
the current state may be unavoidable, but the agent still has to make a decision.

3. Goal-Based Agents

The action taken by these agents depends on the distance from their goal (Desired
Situation). The actions are intended to reduce the distance between the current state and
the desired state. To attain its goal, it makes use of the search and planning algorithm.
One drawback of Goal-Based Agents is that they don’t always select the most optimized
path to reach the final goal.

For example

A a goal-based agent’s behavior can easily be changed to go to a different destination,

simply by specifying that destination as the goal. The reflex agent’s rules for when to turn
and when to go straight will work only for a single destination; they must all be replaced
to go somewhere new.

4. Utility Agents

The action taken by these agents depends on the end objective, so they are called Utility
Agents. Utility Agents are used when there are multiple solutions to a problem, and the
best possible alternative has to be chosen. The alternative chosen is based on each state’s
utility. They perform a cost-benefit analysis of each solution and select one that can
achieve the minimum cost goal.

For examples

Consider two drones G and U, where G is a goal-based and U a utility-based agent. (The
two drones have onboard computerized chips, so there is no need for ground control).
These drones are sent on a mission and they have a goal. Both drones detect the given
goal, but G does not know which of its available actions is more efficient or effective.
However, U, based on its utility function, can select the most efficient or effective action.

5. Learning Agents

Learning Agents have learning abilities so that they can learn from their past experiences.
These types of agents can start from scratch and, over time, can acquire significant
knowledge from their environment. The learning agents have four major components
which enable them to learn from their experience.

● Critic: The Critic evaluates how well is the agent performing vis-à-vis the set
performance benchmark.
● Learning Elements: It takes input from the Critic and helps agents improve
performance by learning from the environment.
● Performance Element: This component decides on the action to be taken to
improve the performance.
● Problem Generator: Problem Generator takes input from other components and
suggests actions resulting in a better experience.

For example, if the automated taxi exerts a certain braking pressure when driving on a
wet road, then it will soon find out how much deceleration is actually achieved, and
whether it skids off the road. The problem generator might identify certain parts of the
model that are in need of improvement and suggest experiments, such as trying out the
brakes on different road surfaces under different conditions.
3. A. Differentiate between informed and uninformed search.

Ans:-

S.N. Informed Search Uninformed Search

1. It uses knowledge for the searching It doesn’t use knowledge for the
process. searching process.

2. It finds solutions more quickly. It finds solutions slow as compared

to informed search.

3. Cost is low. Cost is high.

4. It may or may not be complete. It is always complete.

5. It consumes less time. It consumes moderate time.

6. It provides the direction regarding No suggestion is given regarding

the solution. the solution in it.

7. It is less lengthy while It is more lengthy while

implementation. implementation.

8. Search, A* Search, Graph Search Depth First Search, Breadth First

Search
B. Consider the following map

i) With explanation, find the optimal cost to get from Sibiu to Bucharest
using Uniform Cost Search.

The algorithm continues on, choosing Pitesti for expansion next and adding a
second path to Bucharest with cost 80+97+101 = 278. It has a lower cost, so it
replaces the previous path and is added to the frontier. It turns out this node now
has the lowest cost, so it is considered next, found to be a goal, and returned. Note
that if we had checked for a goal upon generating a node rather than when
expanding the lowest-cost node, then we would have returned a higher-cost path
(the one through Fagaras).

The complexity of uniform-cost search is characterized in terms of the cost of the

optimal solution, and a lower bound on the cost of each action, with .
Then the algorithm’s worst-case time and space complexity is
which can be much greater than This is because uniform-cost search can explore
large trees of actions with low costs before exploring paths involving a high-cost
and perhaps useful action. When all action costs are equal,

is just and uniform-cost search is similar to breadth-first

search.

Uniform-cost search is complete and is cost-optimal, because the first solution it

finds will have a cost that is at least as low as the cost of any other node in the
frontier. Uniform-cost search considers all paths systematically in order of
increasing cost, never getting caught going down a single infinite path (assuming
that all action costs are > 𝑒 >0).