VQ-VAE
VQ-VAE (Vector Quantized Variational Autoencoder) is a generative model architecture that combines discrete latent representations with variational autoencoders, enabling efficient learning and synthesis of complex data like images and audio.
Generative Models
GET3D (generative 3D model)
GET3D is a generative 3D model designed to produce high-quality three-dimensional objects from 2D images. It leverages advances in machine learning to generate detailed 3D shapes and textures, enabling applications in graphics, gaming, and virtual reality.
Wake–sleep algorithm
The Wake–sleep algorithm is a computational method used in machine learning to train probabilistic generative models, particularly useful in unsupervised learning. It alternates between wake and sleep phases to refine model parameters and recognition networks.
Diffsound (discrete diffusion for audio)
Diffsound (discrete diffusion for audio) is a method in generative audio modeling that applies discrete diffusion processes to create or transform audio data. It is part of a broader class of diffusion-based generative models adapted specifically for the discrete, sequential nature of audio signals.
Inception score (IS)
The Inception score (IS) is a quantitative metric used to evaluate the quality of images generated by generative models, particularly generative adversarial networks (GANs). It measures both the diversity and clarity of generated images by leveraging the output of a pretrained Inception network.
Denoising diffusion probabilistic model (DDPM)
Denoising diffusion probabilistic models (DDPM) are a class of generative models that produce data by reversing a gradual noising process. They have gained prominence for their ability to generate high-quality images and other complex data.
Deep belief network
A deep belief network (DBN) is a generative graphical model composed of multiple layers of stochastic, latent variables. It is commonly used in machine learning for unsupervised feature learning, dimensionality reduction, and classification tasks.
Generative adversarial network (GAN)
Generative Adversarial Networks (GANs) are a class of machine learning frameworks designed to generate new data instances that resemble a given training dataset.
FastRAG (efficient retrieval-augmented generation)
FastRAG is a method in natural language processing that enhances the efficiency of retrieval-augmented generation models by optimizing the way external information is retrieved and integrated during text generation. It aims to improve speed and scalability in applications requiring real-time or large-scale knowledge retrieval.