Scribd
Upload
17 views1 page

Lab End Sem - Deep Learning Using Python12536

The document outlines a lab-end semester exam for a deep learning course using Python, consisting of two questions. The first question involves implementing a Variational Autoencoder (VAE) to generate handwritten digits from the MNIST dataset, while the second question focuses on training a Deep Q-Network (DQN) agent to play the CartPole game. Both tasks require visualization and analysis of the results and performance metrics.

Uploaded by

balajibk333
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
17 views1 page

Lab End Sem - Deep Learning Using Python12536

The document outlines a lab-end semester exam for a deep learning course using Python, consisting of two questions. The first question involves implementing a Variational Autoencoder (VAE) to generate handwritten digits from the MNIST dataset, while the second question focuses on training a Deep Q-Network (DQN) agent to play the CartPole game. Both tasks require visualization and analysis of the results and performance metrics.

Uploaded by

balajibk333
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/ 1

22DS6216OJ DEEP LEARNING USING PYTHON

LAB-END SEM EXAM

(Answer All Questions)

Question 1: Generating Handwritten Digits with a Variational Autoencoder (VAE)

(25 marks)

 Task: Implement a VAE using TensorFlow/Keras to generate new handwritten digit


images similar to those in the MNIST dataset. Train the VAE on the MNIST dataset
and then use the decoder part of the model to generate new images from a latent
space. Visualize the generated images and discuss the quality and diversity of the
generated samples.

Question 2: Training a Deep Q-Network (DQN) Agent to Play CartPole

(25 marks)

 Task: Implement a Deep Q-Network (DQN) agent using PyTorch to play the
CartPole game in the OpenAI Gym environment. Train the DQN agent to balance the
pole for as long as possible. Visualize the agent's performance over time (e.g., total
reward per episode) and analyze the impact of different hyperparameters (e.g.,
learning rate, exploration-exploitation balance) on the agent's learning process.

You might also like