1

OVERVIEW OF ARTIFICIAL INTELLIGENCE

Definition of AI
Artificial Intelligence(AI) is that branch of computer science that creates intelligent
machines that think and act like humans. It is the technology that allows systems to
replicate human behavior and thoughts.

It is one of the revolutionizing technologies that people are fascinated by because of its
ability to relate to their daily lives. AI enables machines to think, learn and adapt, to
enhance and automate tasks across industries.
Applications
 Health care − AI in healthcare is used to assist in tasks like diagnosing diseases,
personalizing treatments and drug discovery.
 Finance − AI is used for fraud detection, trading and stock market analysis and
customer service through chatbots.
 Manufacturing and Industries − AI optimizes production processes, improves
quality and identifies machinery failure.
 Agriculture − AI helps combine technology with agriculture by analyzing soil
conditions.
 Transportation − AI helps in designing autonomous vehicles. Some other tasks
include traffic management and route optimization in maps.
 Customer Service − Chatbots and Virtual assistants are AI applications to improve
user engagement.
 Entertainment and Media − AI helps in content creation, personalized content
recommendations, and target advertising.
 Safety and Security − AI enhances threat detection and automates security
measures.
 etc

Areas Borrowed
AI is a field that combines various scientific and technological disciplines, which
include Computer Science, Biology, Psychology, Linguistics, Mathematics, and
Engineering.

AI Research Areas
 Gaming − AI plays a crucial role in strategic games such as chess, poker, tic-tac-
toe, etc., where machine can think of large number of possible positions based on
heuristic knowledge.
 Natural Language Processing − It enables machines to interact with humans in
natural language.
 Expert Systems − It is an AI based software that enables decision-making ability
similar to a human expert.
 Computer Vision − These systems understand, interpret, and comprehend visual
input on the computer.
 Speech Recognition − Some intelligent systems are capable of hearing and
comprehending the language in terms of sentences and their meanings while a human
 2

talks to it. It can handle different accents, slang words, noise in the background,
change in human’s noise due to cold, etc.
 Handwriting Recognition − The handwriting recognition software reads the text
written on paper by a pen or on screen by a stylus. It can recognize the shapes of the
letters and convert it into editable text.
 Intelligent Robots − Robots are able to perform the tasks given by a human. They
have sensors to detect physical data from the real world such as temperature,
movement, and sound. They have efficient processors, and huge memory, to exhibit
intelligence. In addition, they are capable of learning from their mistakes and they
can adapt to new environment.
 Neural Networks - Robots are aimed at manipulating the objects by perceiving,
picking, moving, modifying the physical properties of object, destroying it, or to
have an effect thereby freeing manpower from doing repetitive functions without
getting bored, distracted, or exhausted.

Challenges With Implementing AI Systems

The main challenges in implementing AI includes −
 Data Quality and Accessibility − AI requires large, high-quality, and relevant
datasets for effective learning.
 Technical Expertise − Implementing AI algorithms and models requires skilled
professionals.
 Ethical and Legal Concerns − It is important to make sure that the AI systems are
fair, unbiased, and don't harm anyone's safety.
 Integration − Integrating AI with existing systems can be complex.
 Cost − Developing and maintaining of AI infrastructure can be expensive.

Conventional Programs Versus AI Programs

Conventional Programs (e.g. C) AI programs (e.g. Prolog)
Information, processing usually combined Knowledgebase clearly separate from the
in one sequential program. processing (inference)
Program does not make mistakes Program may make mistakes
Do not (usually) explain why input data is Explanation is part of most expert systems
needed or why conclusions reached
Require all input data. May not function Do not require all initial facts. Typically
properly with missing data can arrive at reasonable conclusion with
missing facts
Changes in program are tedious Changes in rules are easy to make
Program operates only when complete Program can operate with only a few rules
Execution done step by step Execution done by using rules (heuristics)
and logic
Effective manipulation of large databases Effective manipulation of large knowledge
bases
Representation and use of data Representation and use of knowledge
Efficiency is the major goal (e.g. program Effectiveness is the major goal (e.g. how
should be as small as possible) good an advice/conclusion is)
Easily deal with quantitative data Easily deal with qualitative data
 3

(relationships) e.g. identifying patterns

Use numerical data representations Use symbolic knowledge representations

Thus, Key Things with AI Programs

 Use of facts and rules (heuristics) as knowledgebase.
 Symbolic representation of things.
 Qualitative data that has relationships.
 Pattern recognition.

What is Intelligence Composed of?

The intelligence is intangible. It is composed of −
 Reasoning
 Learning
 Problem Solving
 Perception
 Linguistic Intelligence

Reasoning − It is the set of processes that enables us to provide basis for judgement,
making decisions, and prediction. There are broadly two types −

Inductive Reasoning Deductive Reasoning

It starts with a general statement and

It conducts specific observations to makes
examines the possibilities to reach a
broad general statements.
specific, logical conclusion.
Even if all of the premises are true in a If something is true of a class of things in
statement, inductive reasoning allows for general, it is also true for all members of
the conclusion to be false. that class.
Example − "All women of age above 60
Example − "Nita is a teacher. Nita is
years are grandmothers. Shalini is 65
studious. Therefore, All teachers are
years. Therefore, Shalini is a
studious."
grandmother."

Learning − It is the activity of gaining knowledge or skill by studying, practising, being

taught, or experiencing something. Learning enhances the awareness of the subjects of
the study. The ability of learning is possessed by humans, some animals, and AI-enabled
systems.

Problem Solving − It is the process in which one perceives and tries to arrive at a desired
solution from a present situation by taking some path, which is blocked by known or
unknown hurdles. Problem solving also includes decision making, which is the process
of selecting the best suitable alternative out of multiple alternatives to reach the desired
goal are available.
 4

Perception − It is the process of acquiring, interpreting, selecting, and organizing

sensory information. Perception presumes sensing. In humans, perception is aided by
sensory organs. In the domain of AI, perception mechanism puts the data acquired by the
sensors together in a meaningful manner.

Linguistic Intelligence − It is one’s ability to use, comprehend, speak, and write the
verbal and written language. It is important in interpersonal communication.

AI Development Frameworks
There exists pre-built tools and resources that simplify developing AI-based
applications. Some of the top AI frameworks and libraries include;

PyTorch
PyTorch is an open-source framework based on theTorch library and is widely used for
applications in deep learning and artificial intelligence. It provides a flexible and dynamic
computational graph, which makes it a popular choice. Developers use this for various
tasks like Computer Vision and Natural Language Processing.
PyTorch is commonly used for building deep learning models, and applications
like image recognition and language processing.

Scikit-Learn
Scikit-Learn is an open-source library in Python programming language. It simplifies the
process of building and deploying Machine Learning models and algorithms. It is an
user-friendly interface and has a comprehensive range of tools, especially for Data
Mining and deep learning tasks.
Scikit-learn is primarily used for performing tasks like classification, regression,
clustering, dimensionality reduction, feature selection, and data preprocessing.

TensorFlow
TensorFlow is an open-source deep learning framework developed by Google. It is
flexible and scalable, and often used by developers to build and train machine learning
models. It is well-documented and supports deployment on various platforms.
TensorFlow is used for developing machine learning models like image recognition,
handwriting recognition, object detection, sentiment analysis, and machine
translation.

Keras
Keras is an open-source high-level Neural Networks API that runs top of the TensorFlow
Library and other frameworks. It is easy to learn and is user-friendly and is usually used
for building and training deep learning models.

Microsoft Cognitive Toolkit

Microsoft Cognitive ToolKit (CNTK) is an open-source deep learning framework
developed by Microsoft. It is designed to train deep neural networks and offers a wide
range of features and capabilities, and supports multiple neural network types, including
feedforward and recurrent networks.
 5

CNTK is used to create machine learning prediction models, and also create deep
neural networks, such as Cortana and self-driving cars.

LangChain
LangChain is one of the popular frameworks for large language model (LLM)
applications. It integrates with various tools like OpenAI and Hugging Face
Transformers and is used for many applications like chatbots, document
summarization, and interacting with APIs.
LangChain allows developers to chain together tasks like data retrieval, processing,
and LLM calls in a sequential manner.