Machine Learning

What is Machine Learning?

What is Machine Learning?
 In general, Artificial Intelligence, Machine Learning, and Deep Learning
are related as follows:
 “Deep Learning is a kind of Machine Learning, and Machine Learning is a
kind of Artificial Intelligence.”
What is Machine Learning?

What is Learning ?
Learning from Experiences

Learning from Data

What is Machine Learning?
Machine Learning
It is an application of artificial intelligence (AI) that provides systems
the ability to automatically learn and improve from experience
without being explicitly programmed.

ML is based on the idea that systems can learn from data and make
decisions with minimal human intervention.
Data Training
Model

Cat
Answer
Questions Predictions
Dog
ML using Data to answer questions.
Machine Learning is the technique used to find ( or learn ) a model from the data.
It is suitable for problems that involve intelligence, such as image recognitionand
speech recognition, where physical laws or mathematical equations failto produce
a model.

The quality or quantity data available for training and testing play a significant role
in determining the performance of a machine-learning model (Affected to the
learning and prediction performance).
Data in Machine Learning
 Data is a crucial component in the field of Machine Learning.
 Data can be in various forms such as numerical, categorical, or
time-series data, and can come from various sources such as
databases, spreadsheets, or APIs.
 Machine learning algorithms use data to learn patterns and
relationships between input variables and target outputs, which
can then be used for prediction or classification tasks.
Data in Machine Learning
 How do we split data in Machine Learning?

Data in Machine Learning

Training Data Validation Data Testing Data

Data in Machine Learning
 How do we split data in Machine Learning?

 Training Data: The part of data we use to train our model. This is
the data that your model actually sees(both input and output)
and learns from.
Data in Machine Learning
 How do we split data in Machine Learning?

 Validation Data: The part of data that is used to do a frequent

evaluation of the model, fit on the training dataset along with
improving involved hyperparameters (initially set parameters
before the model begins learning). This data plays its part when
the model is actually training.
Data in Machine Learning
 How do we split data in Machine Learning?
 Testing Data: Once our model is completely trained, testing data provides
an unbiased evaluation. When we feed in the inputs of Testing data, our
model will predict some values(without seeing actual output). After
prediction, we evaluate our model by comparing it with the actual output
present in the testing data. This is how we evaluate and see how much our
model has learned from the experiences feed in as training data, set at the
time of training.
Challenges:
 Challenges with ML
Machine Learning is the technique used to find ( or learn ) a model from the data. It is

suitable for problems that involve intelligence, such as image recognition and speech

recognition, where physical laws or mathematical equations fail to produce a model.

 Challenges with ML
 Once the Machine Learning process finds the model from the training data, we

apply the model to the actual field data.

 No Machine Learning approach can achieve the desired goal with the wrong

training data.

 The process used to make the model performance consistent regardless of

the training data or the input data is ca lled generalization.

 Generalization means model with unbiased training data.

 One of the primary causes of corruption of the generalization process is overfitting.

Overfitting occurs when the model has been overly customized to the training data that
it yields poor performance for the actual input data, while its performance for the training
data is excellent.

Overfitting is one of the primary factors that reduces the generalization performance.
Types of Learning
Types of Machine Learning
 Supervised Learning
 The training set given for supervised learning is the labeled dataset.
 Supervised learning tries to find the relationships between the feature set
and the label set, which is the knowledge and properties we can learn
from labeled datasets.
Types of Machine Learning
 Supervised Learning
Given: training data + desired outputs (labels)

 Learning mapping between input X and desired output Y

 Teacher gives me Y’s for the learning purposes
 Types of Machine Learning
 Unsupervised learning
 The training set given for unsupervised learning is the unlabeled dataset
 Unsupervised learning is generally used for investigating the characteristics
of the data and preprocessing the data.
 Types of Machine Learning
 Unsupervised learning

Learning relations between data components

No specific outputs given by a teacher
 Types of Machine Learning
 Reinforcement learning
 Reinforcement learning is used to solve problems of decision-making
(usually a sequence of decisions), such as robot perception and
movement, automatic chess player, and automatic vehicle driving.

 Learning mapping between input X and desired output Y

 Critic does not give me Y’s but instead a signal (reinforcement) of
how good my answer was
 Types of Machine Learning
 Reinforcement learning
Rewards from a sequence of actions
learning from interactions with the environment (experiences).
Use of Machine Learning
 Predictive modeling: Machine learning can be used to build predictive
models that can predict future outcomes based on historical data. This can
be used in many applications, such as stock market prediction, fraud
detection, weather forecasting, and customer behavior prediction.
 Image recognition: Machine learning can be used to train models that can
recognize objects, faces, and other patterns in images. This is used in
many applications, such as self-driving cars, facial recognition systems, and
medical image analysis.
Use of Machine Learning
 Natural language processing: Machine learning can be used to analyze
and understand natural language, which is used in many applications, such
as chatbots, voice assistants, and sentiment analysis.
 Recommendation systems: Machine learning can be used to build
recommendation systems that can suggest products, services, or content
to users based on their past behavior or preferences.
Use of Machine Learning
 Data analysis: Machine learning can be used to analyze large datasets and
identify patterns and insights that would be difficult or impossible for
humans to detect.
 Robotics: Machine learning can be used to train robots to perform
tasks.
 Issues of using data in Machine Learning
 Data quality: One of the biggest issues with using data in machine
learning is ensuring that the data is accurate, complete, and
representative of the problem domain. Low-quality data can result in
inaccurate or biased models.
 Data quantity: In some cases, there may not be enough data available to
train an accurate machine-learning model. This is especially true for
complex problems that require a large amount of data to accurately
capture all the relevant patterns and relationships.
 Issues of using data in Machine Learning
 Bias and fairness: Machine learning models can sometimes perpetuate
bias and discrimination if the training data is biased or unrepresentative.
This can lead to unfair outcomes for certain groups of people, such as
minorities or women.
 Overfitting and Underfitting: Overfitting occurs when a model is too
complex and fits the training data too closely, resulting in poor
generalization to new data. Underfitting occurs when a model is too
simple and does not capture all the relevant patterns in the data.
 Issues of using data in Machine Learning
 Privacy and security: Machine learning models can sometimes be used to
infer sensitive information about individuals or organizations, raising
concerns about privacy and security.
 Interpretability: Some machine learning models, such as deep neural
networks, can be difficult to interpret and understand, making it
challenging to explain the reasoning behind their predictions and
decisions.