Deep & Reinforcement Learning Unit 1

CS702(B) Unit 1 Deep Learning Fundamentals study material for RGPV CSE 7th Semester. Learn history of deep learning, McCulloch Pitts neuron, thresholding logic, activation functions, gradient descent optimizers, RNN, BPTT, GRU, LSTM, encoder-decoder models and attention mechanism.

View Topics Resources Questions

Unit 1 Overview

Unit 1 introduces the foundation of Deep Learning. It covers artificial neuron models, activation functions, optimization algorithms and sequence learning models such as RNN, GRU, LSTM, encoder-decoder architecture and attention mechanism.

🧠

Deep Learning Basics

Understand history, neuron models, thresholding logic and activation functions.

📉

Optimization Methods

Learn Gradient Descent, Momentum, Nesterov, AdaGrad, RMSProp and Adam.

🔁

Sequence Models

Study RNN, BPTT, GRU, LSTM, encoder-decoder and attention mechanism.

Unit 1 Topics Covered

Complete syllabus-based topics of Deep & Reinforcement Learning Unit 1.

History of Deep Learning

Deep Learning evolved from artificial neural networks and became powerful due to big data, better hardware and improved training algorithms.

McCulloch Pitts Neuron

McCulloch Pitts neuron is an early mathematical model of an artificial neuron that uses binary inputs and threshold logic.

Thresholding Logic

Thresholding logic activates a neuron only when the weighted sum of inputs crosses a fixed threshold.

Activation Functions

Activation functions introduce non-linearity into neural networks. Common examples include Sigmoid, Tanh, ReLU and Softmax.

Gradient Descent

Gradient Descent is an optimization algorithm used to minimize loss by updating weights in the opposite direction of the gradient.

Momentum Based Gradient Descent

Momentum-based GD speeds up learning by using past gradients to smooth the update direction.

Nesterov Accelerated Gradient

Nesterov Accelerated Gradient improves momentum by looking ahead before calculating the gradient.

Stochastic Gradient Descent

SGD updates model weights using one or a small batch of training examples at a time.

AdaGrad

AdaGrad adapts the learning rate for each parameter based on past gradients.

RMSProp

RMSProp controls the learning rate using a moving average of squared gradients.

Adam Optimizer

Adam combines ideas of momentum and RMSProp to provide efficient adaptive optimization.

Eigenvalue Decomposition

Eigenvalue decomposition is a linear algebra technique useful in data transformation, dimensionality reduction and model analysis.

Recurrent Neural Networks

RNNs are neural networks designed for sequential data by maintaining memory of previous inputs.

Backpropagation Through Time

BPTT is used to train RNNs by unfolding the network over time and applying backpropagation.

Vanishing and Exploding Gradients

These problems occur in deep or recurrent networks when gradients become too small or too large during training.

Truncated BPTT

Truncated BPTT reduces training complexity by backpropagating errors only for a limited number of time steps.

GRU

Gated Recurrent Unit is an RNN variant that uses gates to control information flow and reduce vanishing gradient problems.

LSTM

Long Short-Term Memory networks use memory cells and gates to learn long-term dependencies in sequence data.

Encoder Decoder Models

Encoder-decoder models convert input sequences into internal representations and generate output sequences.

Attention Mechanism

Attention mechanism helps models focus on important parts of input sequences while generating output.

Quick Revision

Deep Learning: Neural network-based learning with multiple layers.

Gradient Descent: Loss ko minimize karne ke liye weights update karna.

RNN: Sequential data ke liye neural network with memory.

LSTM/GRU: Improved RNN models jo long-term dependencies handle karte hain.

Attention: Model ko input ke important parts par focus karne me help karta hai.

Download Study Resources

📘

Detailed Notes

Download Notes

Important Questions

Download Questions
📄

PYQ Analysis

Download PYQ

Important Questions

  1. Explain history and evolution of Deep Learning.
  2. Explain McCulloch Pitts neuron model.
  3. Explain thresholding logic.
  4. Explain different activation functions used in deep learning.
  5. Explain Gradient Descent algorithm.
  6. Differentiate between GD and SGD.
  7. Explain Momentum Based Gradient Descent.
  8. Explain Nesterov Accelerated Gradient.
  9. Explain AdaGrad, RMSProp and Adam optimizers.
  10. Explain Eigenvalue Decomposition.
  11. Explain Recurrent Neural Networks.
  12. Explain Backpropagation Through Time.
  13. Explain vanishing and exploding gradient problems.
  14. Explain Truncated BPTT.
  15. Explain GRU architecture.
  16. Explain LSTM architecture and its gates.
  17. Differentiate between GRU and LSTM.
  18. Explain Encoder Decoder model.
  19. Explain Attention Mechanism.
  20. Write short note on attention over images.

PYQ Analysis Table

Topic Expected Frequency Importance
McCulloch Pitts Neuron High ⭐⭐⭐⭐
Activation Functions Very High ⭐⭐⭐⭐⭐
Gradient Descent Very High ⭐⭐⭐⭐⭐
AdaGrad, RMSProp, Adam Very High ⭐⭐⭐⭐⭐
RNN Very High ⭐⭐⭐⭐⭐
BPTT High ⭐⭐⭐⭐
Vanishing and Exploding Gradients Very High ⭐⭐⭐⭐⭐
GRU High ⭐⭐⭐⭐
LSTM Very High ⭐⭐⭐⭐⭐
Attention Mechanism Very High ⭐⭐⭐⭐⭐

FAQs

What is Deep Learning?

Deep Learning is a machine learning technique that uses multiple layers of neural networks to learn complex patterns from data.

What is Gradient Descent?

Gradient Descent is an optimization algorithm used to minimize loss by updating model weights.

What is RNN?

RNN is a neural network used for sequential data by maintaining memory of previous inputs.

What is LSTM?

LSTM is an improved RNN model that can learn long-term dependencies using gates and memory cells.

What is Attention Mechanism?

Attention mechanism allows a model to focus on important parts of input while generating output.