# Non-Adversarial Image Synthesis with Generative Latent Nearest Neighbors

## TL;DR

This paper introduces GLANN, a new non-adversarial image generation method that overcomes GAN limitations like instability and mode collapse, achieving superior results by combining IMLE and GLO techniques.

## Contribution

GLANN is a novel approach that combines IMLE and GLO to enable stable, non-adversarial image synthesis with improved quality over existing methods.

## Key findings

- GLANN outperforms 800 GANs and VAEs on standard datasets.
- The method avoids mode collapse and training instability.
- GLANN is effective for non-adversarial unsupervised image translation.

## Abstract

Unconditional image generation has recently been dominated by generative adversarial networks (GANs). GAN methods train a generator which regresses images from random noise vectors, as well as a discriminator that attempts to differentiate between the generated images and a training set of real images. GANs have shown amazing results at generating realistic looking images. Despite their success, GANs suffer from critical drawbacks including: unstable training and mode-dropping. The weaknesses in GANs have motivated research into alternatives including: variational auto-encoders (VAEs), latent embedding learning methods (e.g. GLO) and nearest-neighbor based implicit maximum likelihood estimation (IMLE). Unfortunately at the moment, GANs still significantly outperform the alternative methods for image generation. In this work, we present a novel method - Generative Latent Nearest Neighbors (GLANN) - for training generative models without adversarial training. GLANN combines the strengths of IMLE and GLO in a way that overcomes the main drawbacks of each method. Consequently, GLANN generates images that are far better than GLO and IMLE. Our method does not suffer from mode collapse which plagues GAN training and is much more stable. Qualitative results show that GLANN outperforms a baseline consisting of 800 GANs and VAEs on commonly used datasets. Our models are also shown to be effective for training truly non-adversarial unsupervised image translation.