Introduction to

Artificial Intelligence

by Astrid Wahyu
Definition of AI from John
McCarthy (1956)
Artificial Intelligence (AI) is the science and engineering of making intelligent machines that can
perform tasks that would typically require human intelligence. These tasks include learning,
reasoning, problem-solving, perception, and language understanding. In order to achieve this, AI
systems need to be able to learn from data and improve their performance over time, reason about
complex problems to make decisions, and self-correct when errors are found in their output.
The Learning Process in AI
One of the key components of Artificial Intelligence (AI) is the ability to learn from data. The learning
process in AI involves training a model on a dataset, which is a collection of examples that represent
the problem at hand. The model learns from the dataset by adjusting its parameters to minimize the
difference between its predictions and the actual outcomes. This is known as the "training" phase.

After the model has been trained, it can be used to make predictions on new data that it hasn't seen
before. This is known as the "testing" phase. The model's performance on the testing data will give an
indication of how well it has learned the underlying patterns in the data.

There are several types of learning in AI, including supervised learning, unsupervised learning, and
reinforcement learning. Each type of learning has its own set of techniques and algorithms that are
used to train models on different types of data.
The Reasoning Process in AI
Artificial Intelligence (AI) systems are designed to perform complex reasoning tasks that involve
logical thinking, problem-solving, and decision-making. The reasoning process in AI involves using
existing knowledge, rules, and inference techniques to draw conclusions, make inferences, and solve
problems.
One of the key techniques used in AI reasoning is inference. Inference refers to the process of deriving
new information or conclusions from existing knowledge or evidence. It involves applying logical
rules or algorithms to make logical deductions or draw probable conclusions based on the available
data.
The Reasoning Process in AI
Inference techniques in AI can vary depending on the specific problem domain and the type of
reasoning required. Some common inference techniques include:

• Forward Chaining: In this technique, AI systems start with the available data and use logical
rules to derive new information step by step until a conclusion is reached.
• Backward Chaining: In contrast to forward chaining, this technique starts with the desired
conclusion and works backward, using logical rules to determine the evidence or facts
needed to support the conclusion.
• Rule-Based Reasoning: This technique involves using a set of predefined rules and logical
conditions to make inferences and draw conclusions. The AI system matches the available
data with the rules to determine the appropriate actions or decisions.
• Probabilistic Reasoning: In scenarios where uncertainty is involved, probabilistic reasoning
techniques are used. These techniques assign probabilities to different outcomes and use
statistical methods to make inferences based on the available data.

The use of inference techniques in AI allows systems to reason and make decisions based on logic
and evidence, enabling them to solve complex problems and provide intelligent solutions.
The Self-Correction Process in AI
During the self-correction process, AI models continuously evaluate their predictions and compare
them to the actual outcomes. If the model's predictions are incorrect, it can learn from these
mistakes and update its internal parameters to improve future predictions.

This process often involves using optimization algorithms, such as gradient descent, to iteratively
adjust the model's parameters based on the errors observed during training or testing. By minimizing
the difference between predicted and actual outcomes, the model becomes more accurate and
reliable.
The self-correction process is particularly important in machine learning, where models learn from
large datasets. Through continuous learning and refinement, AI systems can adapt to changing
conditions and improve their performance over time.
Approaches to Artificial
Intelligence
When it comes to developing Artificial Intelligence (AI) systems, there are different approaches to
consider. These approaches include:

1. Thinking Humanly: This approach focuses on developing AI systems that mimic human
thought processes and behaviors. It aims to understand and replicate human cognitive
abilities, such as perception, learning, and problem-solving.
2. Acting Humanly: This approach aims to create AI systems that can mimic human behavior
and interactions. It focuses on developing systems that can communicate, understand, and
respond to humans in a way that is indistinguishable from a human.
3. Thinking Rationally: This approach emphasizes the development of AI systems that follow
logical and rational thought processes. It focuses on using formal rules and algorithms to
solve problems and make decisions based on logical reasoning.
4. Acting Rationally: This approach aims to create AI systems that make decisions and take
actions that maximize their chances of achieving their goals. It focuses on creating systems
that can analyze information, evaluate options, and choose the best course of action based
on rational decision-making.

By considering these different approaches, AI researchers and developers can explore various ways to
create intelligent systems that can think, act, and interact in different ways.
How AI Works
Artificial Intelligence (AI) systems work by processing data and using algorithms to make predictions,
solve problems, and perform tasks. The general steps involved in how AI works are as follows:

1. Data Collection: AI systems first need to collect relevant data from various sources. This
data can include text, images, videos, or any other form of information that is necessary for
the AI system to learn and make decisions.
2. Data Preprocessing: Once the data is collected, it needs to be preprocessed to ensure it is
in a suitable format for the AI system to analyze. This step involves cleaning the data,
removing noise, and transforming it into a structured format.
3. Algorithm Selection: Based on the task or problem at hand, an appropriate algorithm is
selected. Different algorithms have different strengths and weaknesses, and the choice of
algorithm depends on the specific requirements of the AI system.
4. Model Training: The selected algorithm is then used to train a model using the
preprocessed data. During training, the model learns patterns, relationships, and rules from
the data to make accurate predictions or perform tasks.
5. Model Evaluation: Once the model is trained, it needs to be evaluated to assess its
performance. This evaluation is done using a separate set of data called the validation set,
which was not used during training.
6. Model Deployment: If the model performs well during evaluation, it can be deployed to
make predictions or perform tasks in real-world scenarios. This can involve integrating the
model into existing systems or creating new applications that utilize the AI capabilities.
7. Continuous Learning: To adapt to changing conditions or improve over time, AI systems
can undergo continuous learning. This involves updating the model with new data,
retraining it, and deploying the updated model.

By following these steps, AI systems can analyze data, learn from it, and make informed decisions or
perform tasks that were previously only possible for humans.
Types of Artificial Intelligence
1. Narrow AI 2. General AI 3. Superintelligent
AI
Designed for a particular Capable of understanding,
task, narrow AI is focused on learning, and applying its An AI that surpasses human
performing specific intelligence across a range intelligence and
functions within set of tasks and contexts. capabilities, often with
parameters. speculative implications.
Applications of Artificial
Intelligence
1 Healthcare 2 Finance 3 Transportation
AI is used for medical AI drives algorithmic AI powers
imaging analysis, drug trading, fraud autonomous vehicles,
discovery, detection, risk traffic management
personalized assessment, and systems, and route
treatment plans, and customer service optimization for
predictive analytics. automation in the logistics.
financial industry.
Advanced AI Applications
Artificial Intelligence has a wide range of applications across various
industries. Some advanced applications include natural language processing
for chatbots, computer vision for autonomous vehicles, recommendation
systems for personalized experiences, and predictive analytics for forecasting
trends and patterns. These applications are revolutionizing the way we live
and work.
Deepface Facebook

One example of AI is the DeepFace technology owned by Facebook. This AI functions to recognize the
faces of people in photo posts. With this technology, users no longer need to manually tag someone
in a photo, because AI does it. AI can identify people in photos by being trained based on data. Data is
obtained when a user tags a person beforehand and from the results of AI suggestions the person in
the photo is approved by the user.
Virtual Assistance
• Answering questions and providing information

• Setting reminders and alarms

• Managing appointments and schedules

• Playing music or podcasts

• Controlling smart home devices

• Providing weather updates

• Ordering products or services online

• Sending messages or making calls

Popular virtual assistants include Apple's Siri, Amazon's Alexa, Google Assistant, and Microsoft's
Cortana. These virtual assistants leverage AI technologies, such as natural language processing and
machine learning, to continuously improve their performance and better understand user
preferences.
Benefits and Challenges of
Artificial Intelligence
Benefits Challenges
Increased efficiency, improved decision- Privacy concerns, job displacement, bias in
making, enhanced accuracy, and new job algorithms, and ethical dilemmas.
creation.
Ethical Considerations in
Artificial Intelligence

Ethics Equity Transparency

Ensuring that AI systems are Fostering diversity and equity Advocating for transparency
developed and used in an in the development and and accountability in AI
ethical and morally deployment of AI decision-making processes.
responsible manner. technologies.
Conclusion

1.5M 75%
AI Jobs by 2025 Enterprises Adopting AI
The growing demand for AI expertise is Around 75% of commercial enterprise
projected to create an estimated 1.5 million applications will utilize AI in some form by 2024.
jobs by 2025.
 Task 1

• This is an individual task.

• Please make a summary of the inference techniques (the definition,
the types, the rules, etc)
• Submit your pdf file through this link: https://bit.ly/task1-ai
• The due date is on Thursday, 21 March 2024
Thank you!