CSIS 273

Personal Computing for

Applied Sciences

Chapter 6
Artificial Intelligence
 Definition of AI
• Artificial Intelligence (AI) refers to a
broad field of science encompassing
not only computer science but also
psychology, philosophy, linguistics and
other areas.
• AI is concerned with getting computers
or computer-controlled robots to do
tasks that would normally require
human intelligence.
 Examples of AI Technologies
• Several examples of AI impact our lives today.
Subsets of AI

Note that these subsets are

often used in combination.
1. Machine learning
• Machine learning (ML) is a subset of AI that focuses on creating systems
or algorithms that can learn from data and improve their performance
without explicit programming or human intervention.
• ML can be divided into three main types:
• Supervised learning: when the system learns from labeled data,
such as images with captions
• Unsupervised learning: when the system learns from unlabeled
data, such as clustering similar items
• Reinforcement learning: when the system learns from its own
actions and feedback, such as playing a game, controlling a robot,
etc.
• AI and ML are closely related and often used interchangeably, but they
are not the same thing. ML is one of the methods or tools that can be
used to achieve AI, but not all AI systems use ML.
Some applications of machine learning

• Spam filtering: analyzing email

messages and classifying them as
spam or not spam.
• Face recognition.
• Recommendation systems:
suggesting products, services, or
content to users based on their
preferences, behavior, and
feedback.
• Self-driving cars.
• Social media monitoring.
• The vast majority of advancements in AI
today are due to machine learning models:
• Google Search was once based solely on
rules written by engineers, which limited the
number of queries it could handle.
• Today, Google Search is powered by machine
learning, which allows it to handle billions of
queries per day and get smarter over time.
• YouTube uses machine learning to
recommend videos to users.
• Amazon uses it to personalize product
recommendations.
• Netflix uses it to provide customized
recommendations for TV shows and movies.
Generative AI

• Until recently, machine learning was largely limited to predictive models,

used to observe and classify patterns in content.
• Analytical AI: A classic machine learning problem starts with an image or
several images (for example: adorable cats).
• The program would then identify patterns among the images, and then
scrutinize random images for ones that would match the adorable cat
pattern.
• Generative AI was a breakthrough.
• Rather than simply perceive and classify a photo of a cat, machine learning
is now able to create an image or text description of a cat on demand.
• So generative AI can generate new data in text, images, video, code and
audio based on what it learns.
a. Image generation

• Image generation can be used in areas like digital art, computer

graphics, medical imaging, or just for fun.
• An earlier iteration of AI-generated images relied on generative
adversarial networks (GANs)
• Newer models take a completely different approach to image
creation called diffusion before they generate a novel image based
on a user’s command.
• Example:
DALL-E from OpenAI
Images close to reality

• Berlin-based Boris Eldagsen won the creative open category at this

year's Sony World Photography Award with his entry "Pseudomnesia:
The Electrician.“
• He rejected the award after revealing that his submission was
generated by (AI).

Read the complete article here:

https://edition.cnn.com/style/article/ai-photo-
win-sony-scli-
intl/index.html#:~:text=Top%20image%3A%20Bo
ris%20Eldagsen's%20AI,in%20the%20creative%2
0open%20category.
b. Video generation

• It can be used to create talking head videos from text for marketing,
training, corporate communications…
• https://youtu.be/wMn-REUq6Ws
c. Text Generation: Chatbots
• a computer program that uses artificial
intelligence (AI) technologies to simulate a
conversation with human users.
• designed to interact with users in natural
language, either through text or voice,
• can be used for a wide range of applications,
including customer service, information retrieval,
and entertainment.
• use natural language processing (NLP) techniques
to understand user queries and generate
appropriate responses.
• can be integrated into websites, messaging
platforms, and mobile apps to provide users with
instant access to information and support.
Examples of chatbots
• ChatGPT:
• an AI chatbot created by OpenAI, released in November 2022.
• built on a family of large language models (LLMs) collectively known as GPT-3.
• These models can understand and generate human-like answers to text
prompts because they've been trained on huge amounts of data.
• can generate new text similar in style and content to a given prompt,
• can summarize a text.
• can also write codes, poems, songs, and even short stories in the style of a
specific author.
• Bing Chatbot: embedded in the Microsoft
Edge browser, powered by ChatGPT.

• Google Bard created by Google.

Chatbots Avatar
• Chatbots can have 3-D avatar.
• t is currently being used by businesses that are looking for virtual asssistants
to provide a unique and personalized experience to their customers.
• Example: Sydney from Bing.

• Ibtikar from Al Jazeera:

https://www.youtube.com/watch?v=0KtKhyX5pTU
2. Natural Language Processing (NLP)
• It is a subfield of artificial intelligence (AI) that focuses on the interaction
between computers and humans in natural language. NLP enables computers
to understand, interpret, and generate human language.
• NLP is used in a wide range of applications:
• language translation,
• sentiment analysis,
• speech recognition,
• text-to-speech conversion,
• text editors and autocorrect (grammarly)
• It is also used in chatbots and virtual assistants (Siri, Alexa..) to enable them
to understand and respond to user queries in natural language.
• Machine learning techniques are often used in NLP to develop models that
can understand and generate human language.
3. Computer vision
• It is a field of artificial intelligence (AI) that focuses on enabling
computers to interpret and understand visual information from the
world.
• It involves the development of algorithms and techniques that can
analyze and understand images and videos.
• Applications:
• image recognition,
• object detection,
• facial recognition,
• autonomous vehicles.
4. Robotics
• Robots are machines that can be programmed to
perform tasks autonomously or semi-
autonomously.
• Robotics has many applications, including
manufacturing, healthcare, transportation, and
space exploration.
• Humanoid robotics is a field of robotics that
focuses on the development of robots that
resemble and act like humans. They are typically
engineered to imitate authentic human
expressions, interactions, and movements.
• Example: Ameca
https://www.youtube.com/watch?v=Z12kNRDgvG
Q
 Benefits of robots
• Humanoid robots are constantly alert,
minimizing the potential for human-like error.
• Humanoid robots can independently navigate
their environment and perform programmed
tasks with minimal human oversight.
• Deploying humanoid robots can help reduce
human exposure to life-threatening risks or
hazards.
• Humanoid robots can be sent in to assess
dangerous situations, such as a fire incident in
a building or a suspicious package, protecting
the lives of guards that must evacuate
immediately, and providing critical information
to first responders.
Advantages of AI
1. Optimizing business processes: such as data analysis, customer service,
and marketing.
2. Saving time and resources: by automating repetitive and tedious tasks,
such as data entry and scheduling.
3. Eliminating human errors and biases: by providing accurate and
consistent results, such as fraud detection and medical diagnosis.
4. Enhancing human capabilities and performance: by augmenting skills,
knowledge, and creativity.
5. Improving customer satisfaction and loyalty: by offering personalized and
tailored services, such as chatbots, recommendations, and feedback systems.
Disadvantages of AI:
1. High costs: Developing and maintaining AI systems can be very expensive
and time-consuming, as they require specialized hardware, software, and
human expertise.
2. Lack of emotion and creativity: AI systems cannot replicate the human
ability to use emotion and creativity in decision-making, which can limit their
applicability in some domains such as art, literature, or social interactions.
3. Degradation over time: AI systems may degrade or malfunction over time
due to wear and tear, bugs, or cyberattacks, which can compromise their
reliability and performance.
4. Reduced jobs for humans: AI systems may replace human workers in
some tasks or sectors, leading to unemployment.
Disadvantages of AI – 2 :
• 5. Ethical problems: AI systems may pose ethical challenges such as:
• The lack of transparency and accountability of AI systems that make
decisions affecting human lives, such as health care, employment, credit,
and justice.
• The ethical implications of using AI to collect, analyze, and manipulate large
amounts of personal data, often without the consent or awareness of the
users.
• The potential for AI to amplify human biases and prejudices, such as
racism, sexism, and discrimination, in its outputs and outcomes.
• The existential risks of creating superintelligent AI that could surpass
human intelligence and control, and pose a threat to humanity's survival
and values.
ChatGPT crisis

• A group of prominent computer scientists and tech leaders have signed a

petition calling for a six-month pause in the development of AI systems
more powerful than GPT-4.
• The petition warns that such AI systems can pose profound risks to society
and humanity, from spreading misinformation and displacing jobs to
threatening human autonomy and control.
• One of the leaders who signed the petition is Elon Musk, a co-founder of
OpenAI. He has criticized an "out-of-control race" to develop and deploy
ever more powerful AI systems that no one can understand, predict or
control.
• He has claimed that ChatGPT has a liberal bias and has announced plans to
create an alternative chatbot called TruthGPT that will be a "maximum
truth-seeking AI“.
• Geoffrey Hinton, the “godfather of AI” just left Google to be able
to talk publicly about the harm of AI.
• For half a century, he worked on developing chatbots like
ChatGPT. Now he worries it will cause serious harm:
1. Microsoft has augmented its Bing search engine with a chatbot
— challenging Google’s core business so Google is racing to deploy
the same kind of technology.
• The tech giants are locked in a competition that might be
impossible to stop.
2. His immediate concern is that the internet will be flooded with
false photos, videos, and text, and the average person will “not be
able to know what is true anymore.”
Read the complete article here:
3. He fears a day when truly autonomous weapons — those killer https://www.nytimes.com/2023/
robots — become reality. 05/01/technology/ai-google-
chatbot-engineer-quits-
• “The idea that this stuff could actually get smarter than people hinton.html

— a few people believed that. But most people thought it was

way off. And I thought it was way off. I thought it was 30 to 50
years or even longer away. Obviously, I no longer think that.”
Downsides of Chatbots
1. Threatening some jobs like: Customer service representatives,
telemarketers, translators, writers, tutors.
2. Privacy concerns and fears of misinformation which lead some countries to
ban ChatGPT.
3. Lack of empathy: they are unable to understand and respond to human
emotions.
4. Inability to understand slang, misspellings, and sarcasm.
5. Non-customized solutions: may provide automatic responses without
considering the needs of each particular user.
6. Weird conversations:
https://www.youtube.com/watch?v=Kza8TF8nb0k
Downsides of humanoid robotics

1. High Cost
2. Limited capabilities
3. Job displacement
4. Ethical concerns
5. Dependance on electricity and maintenance
6. Security risks
7. Social interaction
8. Fear and mistrust
9. Technological limitations
10.Environmental impact

• https://www.youtube.com/watch?v=8qNWwPj5Yd4
5. Neural Networks
• A neural network is a series of algorithms
that endeavors to recognize underlying
relationships in a set of data through a
process that mimics the way the human
brain operates. In this sense, neural
networks refer to systems of neurons,
either organic or artificial in nature.
Neural Networks (Cont)
• Artificial neural networks provide a powerful tool to help doctors
analyze, model, and make sense of complex clinical data across a
broad range of medical applications. Most applications of artificial
neural networks in medicine are classification problems.
• An artificial neural network (ANN) is a computational model that
attempts to account for the parallel nature of the human brain. It
is a network of highly interconnected processing elements
(neurons) operating in parallel. These elements are inspired by
biological nervous systems.
Neural Networks (Cont)
• Medical diagnosis using artificial intelligence (AI) systems,
particularly artificial neural networks and computer-aided
diagnosis with deep learning, is currently a very active research
area in medicine and it is believed that it will be more widely used
in biomedical systems.
• Evolving neural network techniques for medical diagnosis are
broadly considered since they are ideal in recognizing diseases
using scans.
• Neural networks learn by example so the details of how to
recognize the disease are not needed. For instance, the utilization
of deep learning-based ANN models aids in the timely diagnosis of
gastric cancer with sensitivity and specificity.
Biological Neuron vs Artificial Neuron
Biological Neuron
• Neurons use this interconnected network to pass information’s with each
other using electric and chemical signals.
• Although it may seem that neurons are fully connected, two neurons actually
do not touch each other. They are separated by tiny gap call Synapse.
• Each neuron process information and then it can ”connect” to as many as 50
000 other neurons to exchange information. If connection between two
neuron is strong enough (will be explained later) information will be passed
from one neuron to another. On their own, each neuron is not very bright but
put 100 billion of them together and let them talk to each other, then this
system becomes very powerful.
• A typical neuron would have 4 components seen on Figure 1: Dendrites,
Soma, Axon and Synapse. Dendrites gather inputs from other neurons and
when a certain
Synapse in neural network
• A synapse is the connection between nodes, or neurons, in an
artificial neural network (ANN). Similar to biological brains, the
connection is controlled by the strength or amplitude of a connection
between both nodes, also called the synaptic weight.
Artificial Neuron algorithm
• This figure represents the
architecture of an simple NN.
• It is made up from an input, output
and one or more hidden layers.
Each node from input layer is
connected to a node from hidden
layer and every node from hidden
layer is connected to a node in
output layer. There is usually some
weight associated with every
connection.
 Artificial Neuron Algorithm (Cont)

• Activation function can be linear,

threshold or sigmoid function
• SUM is collection of the output nodes
from hidden layer that have been
multiplied by connection weights,
added to get single number and put
through sigmoid function (activation
function).
• Input to sigmoid is any value between
negative infinity and positive infinity
number while the output can only be a
number between 0 and 1.
Artificial Neuron Algorithm (Cont)
• Input layer represents an the raw information that is fed into the
network. This part of network is never changing its values. Every
single input to the network is duplicated and send down to the
nodes in hidden layer.
• Hidden Layer accepts data from the input layer. It uses input
values and modifies them using weight value, this new value is
than send to the output layer but it will also be modified by some
weight from connection between hidden and output layer.
• Output layer process information received from the hidden layer
and produces an output. This output is than processed by
activation function.
Benefits of Deep Neural Networks
• Deep neural networks (DNN) can help interpret medical scans of
pathologies, electrocardiograms, and endoscopy. Particular attention
is paid to radiology, namely, in using neural networks to analyze X-ray
images. Algorithms to interpret chest scans to make 14 different
diagnoses, ranging from pneumonia to cardiac hypertrophy and lung
collapse.
• DNNs can also diagnose certain types of cancer, fractures,
hemorrhages, retinopathy, skin lesions, and many other diseases.
Neural networks algorithms can improve the performance of
dermatologists, cardiologists, ophthalmologists, and even
psychotherapists by tracking the development of depression.
References
• Artificial intelligence (AI) | Definition, Examples, Types .... https://www.britannica.com/technology/artificial-intelligence
Accessed 4/19/2023.
• ChatGPT explained: everything you need to know about the AI chatbot .... https://www.techradar.com/news/chatgpt-
explained Accessed 4/19/2023.
• Artificial Intelligence 101: Understanding the Basics of AI - LinkedIn. https://www.linkedin.com/pulse/artificial-intelligence-
101-understanding-basics-ai Accessed 4/19/2023.
• All the Benefits Of Artificial Intelligence - Western Governors University. https://www.wgu.edu/blog/benefits-artificial-
intelligence2204.html Accessed 4/19/2023.
• The Pros And Cons Of Artificial Intelligence - Forbes. https://www.forbes.com/sites/qai/2022/12/01/the-pros-and-cons-of-
artificial-intelligence/ Accessed 4/19/2023.
• (3) Limitations of AI | SpringerLink. https://link.springer.com/chapter/10.1007/978-1-4842-6206-1_13 Accessed
4/19/2023.
• Will Machines Become More Intelligent Than Humans?. https://scienceexchange.caltech.edu/topics/artificial-intelligence-
research/machines-more-intelligent-than-humans Accessed 4/19/2023.
• https://www.pbs.org/newshour/science/how-ai-makes-images-based-on-a-few-words
• https://www.synthesia.io/post/generative-ai