Scribd
Upload
19 views

Auto Encoder

An autoencoder is an unsupervised neural network that compresses an input into a latent-space representation and then reconstructs the output from this encoding. It contains an encoder that maps the input to a hidden representation and a decoder that maps this representation back to the output. The autoencoder is trained to minimize the reconstruction error between the input and output. There are different types of autoencoders like undercomplete, regularized, sparse, and denoising autoencoders that are used for tasks like dimensionality reduction, data compression, denoising, and more.

Uploaded by

Ethan Sanaaf
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
19 views

Auto Encoder

An autoencoder is an unsupervised neural network that compresses an input into a latent-space representation and then reconstructs the output from this encoding. It contains an encoder that maps the input to a hidden representation and a decoder that maps this representation back to the output. The autoencoder is trained to minimize the reconstruction error between the input and output. There are different types of autoencoders like undercomplete, regularized, sparse, and denoising autoencoders that are used for tasks like dimensionality reduction, data compression, denoising, and more.

Uploaded by

Ethan Sanaaf
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Autoencoder

Autoencoder

• Autoencoder is a feed forward neural network trained by using unsupervised learning.


• Neural networks trained to attempt to copy its input to its output
• It contain two parts:
Encoder: map the input to a hidden representation. An encoder function h = f(x)
Decoder: map the hidden representation to the output (reconstructs the input). The decoder function r = g(h).

2 Architecture
•Encoder: compress input into a latent-space of usually smaller dimension. h = f(x)
•Decoder: reconstruct input from the latent space. r = g(f(x)) with r as close to x as possible

3
Autoencoder

•It basically has three layers: input, hidden (Bottleneck) and output layer

X X’

4
Autoencoder
Traditionally, autoencoders were used for dimensionality reduction or feature learning.

● An Autoencoder can be divided into two parts


● Encoder which maps input into a lower dimensional space and decoder maps back this lower dimensional space into
reconstruction space where dimensionality of reconstructed space is same as input space.
● Eg: take an image, compress it and reconstruct it, original image back.
● In case of Auto encoder it is lossy compression.
● When Auto encoder compresses and uncompresses and when it uncompresses it tries to be closer to the input.
● The difference between input and output representation is called reconstruction error.
● The training objective of Autoencoder is to minimise the error between the input and reconstructed input.
● Reconstruction error is given by |x-x’|. Its also called as loss function.
● During the course of training the model weights are updated to minimise the reconstruction error

Eg: Give 6 pixel values as input to Autoencoder

● Encoder encodes the signal and at hidden layer


six pixels are compressed to 3 pixels.
5 ●
● Further decoder decodes the information (reconstructs the signal) and at output layer 3 pixels are decoded back to 6 pixels.
Autoencoders:
Properties of Autoencoders:
Data-specific: Autoencoders are only able to compress data similar to what they have been trained on.
Lossy: The decompressed outputs will be degraded compared to the original inputs.

Hyperparameters of Autoencoders:
•Code size: It represents the number of nodes in the middle layer. Smaller size results in more compression.

•Number of layers: The autoencoder can consist of as many layers as we want.

•Number of nodes per layer: The number of nodes per layer decreases with each subsequent layer of the encoder, and

increases back in the decoder. The decoder is symmetric to the encoder in terms of the layer structure.

•Loss function: We either use mean squared error or binary cross-entropy. If the input values are in the range [0, 1] then we

typically use cross-entropy, otherwise, we use the mean squared error.

6
Autoencoder types
Undercomplete autoencoders

Regularized autoencoders

Sparse autoencoders

Denoising autoencoder

Contractive autoencoders

7
Undercomplete Autoencoders

• Undercomplete Autoencoders are designed to have


a hidden layer h, with smaller dimension than input layer
x.
• Its designed to extract useful features from data rather
than simply copying input to output.
•Training of Undercomplete autoencoder forces it to
capture silent features of training data.
• Network must model x in lower dim. space + map
latent
8 space accurately back to input space.
Applications

• Denoising: input clean image + noise and train to reproduce the clean image.

• Dimensionality Reduction/Data compression: Make high-quality, low-dimension representation of data


9
Applications

● Water mark removal

● Image colorization: input black and white and train to produce color images

10

You might also like