# Sliced generative models

## TL;DR

This paper introduces sliced generative models based on AutoEncoders that simplify sample discrimination to one dimension, comparing two groups of methods and analyzing their effectiveness in reducing FID scores.

## Contribution

The paper proposes a novel class of AutoEncoder-based generative models utilizing a sliced approach, and compares two different groups of methods for their performance.

## Key findings

- Both groups are valid generative models.
- The second group achieves a slightly faster FID reduction.
- Methods based on classical distances outperform modified normality tests.

## Abstract

In this paper we discuss a class of AutoEncoder based generative models based on one dimensional sliced approach. The idea is based on the reduction of the discrimination between samples to one-dimensional case. Our experiments show that methods can be divided into two groups. First consists of methods which are a modification of standard normality tests, while the second is based on classical distances between samples. It turns out that both groups are correct generative models, but the second one gives a slightly faster decrease rate of Fr\'{e}chet Inception Distance (FID).