Experiments on MNIST, Fashion-MNIST and CIFAR-10 show that our method can significantly improves VAE in terms of FID.

Our proposed VA-VAE (Vision foundation model Aligned Variational AutoEncoder) significantly expands the reconstruction-generation frontier of latent diffusion models, enabling faster convergence of ...

This repository contains implementations of various machine learning algorithms from scratch, including Multi-Layer Perceptron (MLP), Gaussian Mixture Models (GMM), Principal Component Analysis (PCA), ...