Scribd
Upload
4 views

Technical_Report

research on machine learning

Uploaded by

rakib.hosen.tamim.2k2
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Technical_Report

research on machine learning

Uploaded by

rakib.hosen.tamim.2k2
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

An Overview of Machine Learning

Algorithms

December 7, 2024

Abstract
Machine learning has revolutionized the field of artificial intelligence by enabling
systems to learn and improve from experience without being explicitly programmed.
This report provides a concise overview of various machine learning algorithms,
categorizing them into supervised, unsupervised, and reinforcement learning. Key
concepts and representative algorithms for each category are discussed.

1 Introduction
Machine learning (ML) is a subset of artificial intelligence (AI) that focuses on building
systems capable of learning from data. ML algorithms can be broadly categorized into
three types:
• Supervised Learning
• Unsupervised Learning
• Reinforcement Learning

2 Supervised Learning
Supervised learning involves training a model on labeled data. Labeled data means that
we know both the input and the correct output for each example. The model learns to
predict the correct output for new, unseen inputs based on this training data.

2.1 Linear Regression


Linear regression is used to predict a continuous value based on the input features. For
example, if you want to predict house prices based on the house size, the model tries to
find a straight line (or a hyperplane in higher dimensions) that best fits the data.
The equation for linear regression is:
y = mx + c
where: - y is the predicted value, - x is the input, - m is the slope of the line (determined
during training), and - c is the intercept (also learned during training).
Imagine plotting house sizes on the x-axis and house prices on the y-axis. Linear
regression draws a line that passes as closely as possible through all the points.

1
Figure 1: Linear regression: A-line fitting the data to predict continuous values.

2.2 Logistic Regression


Logistic regression is used for classification tasks, where the output is categorical (e.g.,
”yes” or ”no”, ”cat” or ”dog”). Instead of predicting a continuous value, logistic regres-
sion predicts the probability of an outcome.
The model uses a special function called the **sigmoid function**:
1
Sigmoid(z) =
1 + e−z
The sigmoid function ensures that the predicted output is between 0 and 1, representing
a probability. If the probability is above a threshold (e.g., 0.5), the model classifies the
input into one category; otherwise, it assigns the other.
For example, logistic regression can predict whether an email is spam (1) or not spam
(0).

Figure 2: Logistic regression: Using the sigmoid function for classification.

2.3 Support Vector Machines (SVM)


SVM is a powerful algorithm used for both classification and regression tasks. It works
by finding the **best possible boundary** (called a hyperplane) that separates the data
into different categories.

2
For example, imagine you have two types of fruits (e.g., apples and oranges) plotted
on a graph based on their size and color intensity. SVM finds a line (or a curve in higher
dimensions) that maximally separates the two groups.
SVM uses a concept called the **margin**, which is the distance between the hyper-
plane and the nearest data points from each class. The algorithm aims to maximize this
margin, ensuring better separation and robustness.
SVM can also handle cases where data is not linearly separable by using something
called the **kernel trick**, which transforms the data into higher dimensions where a
separation is possible.

Figure 3: Support Vector Machines: Finding the optimal hyperplane for classification.

3 Unsupervised Learning
Unsupervised learning deals with unlabeled data. The goal is to identify patterns or
structures in the data. Common algorithms include:

• Clustering: Techniques such as k-means and DBSCAN group similar data points
together.

• Principal Component Analysis (PCA): Reduces the dimensionality of data


while preserving variance.

Figure 4 provides an example of clustering in unsupervised learning.

Figure 4: An example of unsupervised learning. Clustering groups similar data points


together without labels.

3
4 Reinforcement Learning
Reinforcement learning focuses on training agents to make decisions in an environment
to maximize cumulative rewards. Algorithms such as Q-learning and Deep Q-Networks
(DQN) are widely used in applications like robotics and gaming.
Figure 5 illustrates the reinforcement learning process, where an agent interacts with
the environment.

RL_1_3932c27f5d.jpg

Figure 5: An example of reinforcement learning. The agent learns by interacting with


the environment and receiving rewards or penalties.

5 Applications
Machine learning algorithms have been applied in numerous domains:

• Healthcare: Disease diagnosis and personalized medicine.

• Finance: Fraud detection and algorithmic trading.

• Transportation: Autonomous vehicles and traffic optimization.

4
6 Conclusion
Machine learning is a rapidly evolving field with diverse applications. Understanding
the core algorithms is crucial for leveraging the potential of ML in solving real-world
problems.

References
[1] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press, 2016.

[2] C. M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.

You might also like