Scribd
Upload
2 views4 pages

Machine Learning Deep Learning Complete Notes

This study material provides an overview of Machine Learning and Deep Learning concepts, including types of machine learning, performance evaluation metrics, and the history of deep learning. It covers key topics such as neural networks, optimization techniques like Gradient Descent, and dimensionality reduction methods like PCA and SVD. The document aims to prepare students for their mid-sem exam with clear explanations of essential topics.

Uploaded by

Mahek Gohil
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
2 views4 pages

Machine Learning Deep Learning Complete Notes

This study material provides an overview of Machine Learning and Deep Learning concepts, including types of machine learning, performance evaluation metrics, and the history of deep learning. It covers key topics such as neural networks, optimization techniques like Gradient Descent, and dimensionality reduction methods like PCA and SVD. The document aims to prepare students for their mid-sem exam with clear explanations of essential topics.

Uploaded by

Mahek Gohil
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

Machine Learning & Deep Learning Midsem Study Material

Machine Learning & Deep Learning Midsem Study Material

1. Basic Concepts of Machine Learning and Deep Learning

-----------------------------------------------------

What is Machine Learning?

- Machine Learning is a type of Artificial Intelligence where systems learn from data without explicit

programming.

Types of Machine Learning:

- Supervised Learning: Uses labeled data (Classification, Regression)

- Unsupervised Learning: Uses unlabeled data (Clustering, Association)

- Reinforcement Learning: Learns by rewards and penalties

Performance Evaluation Metrics:

- Confusion Matrix

- Precision = TP / (TP + FP)

- Recall = TP / (TP + FN)

- Accuracy = (TP + TN) / (TP + TN + FP + FN)

- F1-Score = 2 * (Precision * Recall) / (Precision + Recall)

- ROC Curve and AUC

2. Introduction to Deep Learning and Neural Networks

-----------------------------------------------------

What is Deep Learning?


- Deep Learning is a subset of Machine Learning that uses neural networks with multiple layers to

model complex data.

History of Deep Learning:

- 1943: McCulloch-Pitts Neuron (First Mathematical Model)

- 1958: Perceptron by Rosenblatt

- 1980s: Backpropagation Algorithm

- 2012: AlexNet (ImageNet Competition)

Deep Learning Success Stories:

- Image Recognition (AlexNet, ResNet)

- Speech Recognition (Google Assistant, Siri)

- Language Translation (BERT, GPT)

McCulloch-Pitts Neuron:

- Simplified model of biological neurons

- Uses binary inputs and threshold activation function

Multilayer Perceptrons (MLPs):

- Neural networks with input, hidden, and output layers

- Activation Functions: Sigmoid, ReLU

- Can approximate any continuous function

Feed Forward Neural Networks:

- Information flows in one direction (Input -> Hidden -> Output)

- No cycles or loops
Backpropagation:

- Used to train neural networks by updating weights using Gradient Descent

3. Optimization & Encoding

-----------------------------------------------------

Gradient Descent:

- Optimization algorithm to minimize loss function

- Types: Batch Gradient Descent, Stochastic Gradient Descent (SGD), Adam Optimizer

Principal Component Analysis (PCA):

- Dimensionality Reduction Technique

- Transforms correlated features into uncorrelated Principal Components

Singular Value Decomposition (SVD):

- Matrix Factorization Technique

- Used in image compression and NLP

Autoencoders:

- Neural networks used to compress and reconstruct data

- Types:

- Regular Autoencoders

- Denoising Autoencoders (Reconstructs noisy data)

- Sparse Autoencoders (Enforces sparsity to learn features)

Regularization in Autoencoders:

- Prevents overfitting by adding penalties to weights


Conclusion:

This study material covers all essential topics for your mid-sem exam with simple language.

Prepared by ChatGPT | Easy Language Study Material

You might also like