Autoencoder
المشفِّر الذاتي
Definition
A neural network that learns to compress its input into a small latent representation and then reconstruct the input from it.
Definition
An autoencoder is a neural network trained to reconstruct its input through a low-dimensional bottleneck, thereby learning a compact latent representation without labels.
Practical examples
import torch
from torch import nn
ae = nn.Sequential(nn.Linear(8, 2), nn.ReLU(), nn.Linear(2, 8))
x = torch.randn(5, 8)
print(ae(x).shape)Expected output
torch.Size([5, 8])
