Scribd
Upload
34 views3 pages

AI ML DL Notes

The document provides an overview of Artificial Intelligence, focusing on Machine Learning (ML), Deep Learning (DL), and Multilayer Perceptrons (MLP). It outlines the types of ML, key components, applications, and challenges, as well as the differences between ML and DL. Additionally, it details the structure and working of MLP, including its advantages and limitations.

Uploaded by

Jam Falak Shair
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
34 views3 pages

AI ML DL Notes

The document provides an overview of Artificial Intelligence, focusing on Machine Learning (ML), Deep Learning (DL), and Multilayer Perceptrons (MLP). It outlines the types of ML, key components, applications, and challenges, as well as the differences between ML and DL. Additionally, it details the structure and working of MLP, including its advantages and limitations.

Uploaded by

Jam Falak Shair
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Artificial Intelligence: Machine

Learning, Deep Learning & MLP Notes


1. Machine Learning (ML)
Machine Learning is a subfield of Artificial Intelligence (AI) that focuses on enabling
machines to learn from data and make decisions or predictions without being explicitly
programmed for every specific task.

Types of Machine Learning:

- Supervised Learning: The model is trained on labeled data (input-output pairs). It learns to
map inputs to the correct output.
Example: Predicting house prices. Algorithms include Linear Regression, Decision Trees,
SVM.
- Unsupervised Learning: The model is given unlabeled data and it finds patterns or
groupings on its own.
Example: Customer segmentation. Algorithms include K-Means, PCA.
- Reinforcement Learning: The model learns by interacting with an environment and
receiving rewards or penalties.
Example: Game-playing AI. Used in robotics and strategy games.

Key Components of ML:


- Data
- Features
- Model
- Training
- Testing
- Algorithms

Applications:
- Healthcare (disease diagnosis)
- Finance (fraud detection)
- Marketing (targeted advertising)
- Transportation (traffic prediction)
- Natural Language Processing (chatbots)

Challenges:
- Data quality
- Overfitting
- Bias
- Interpretability
2. Deep Learning (DL)
Deep Learning is a specialized subfield of Machine Learning where models learn from large
volumes of data using neural networks with multiple layers. It is inspired by the structure
and function of the human brain.

Core Elements:
- Artificial Neural Networks (ANNs): Input layer, hidden layers, and output layer.

Deep Learning Architectures:


- Feedforward Neural Networks (FNNs): Basic model with data flowing one way.
- Convolutional Neural Networks (CNNs): Ideal for image processing.
- Recurrent Neural Networks (RNNs): Best for sequential data like time series and text.
- Transformers: Powerful in Natural Language Processing (e.g., GPT, BERT).

Working Process:
1. Forward Propagation
2. Loss Calculation
3. Backpropagation
4. Iterative Training

Applications:
- Image and speech recognition
- Natural Language Processing
- Self-driving cars
- Medical imaging

Challenges:
- Requires huge datasets
- High computational cost
- Hard to interpret (black-box models)

3. Difference Between Machine Learning and Deep Learning


Machine Learning and Deep Learning are both subsets of AI, but differ in complexity, data
requirements, and architecture.

Aspect Machine Learning Deep Learning

Data Needs Works with smaller Needs large datasets


datasets

Feature Engineering Manual Automatic

Hardware Less demanding Requires GPUs/TPUs

Accuracy Good for simpler problems High for complex tasks


Algorithms Linear Regression, SVM CNN, RNN, Transformers

Interpretability More transparent Often a black box

4. Multilayer Perceptron (MLP)


A Multilayer Perceptron is a type of neural network that contains multiple layers of
neurons. It is used in supervised learning tasks such as classification and regression.

Structure:
- Input Layer → Hidden Layer(s) → Output Layer

Working:
1. Input is passed through layers (Forward Propagation).
2. Error is calculated (Loss Function).
3. Weights are updated through Backpropagation.

Activation Functions:
- Sigmoid: Binary classification
- ReLU: Fast and widely used
- Tanh: Scales between -1 and 1

Applications:
- Handwriting recognition
- Fraud detection
- Stock prediction
- Medical diagnosis

Advantages:
- Can model complex relationships
- Works for classification and regression tasks

Limitations:
- Computationally intensive
- Risk of overfitting
- Not ideal for image or sequence data (better to use CNN/RNN)

You might also like