Tejkiran·4d agoVariational Autoencoders: From Compressed Representations to Probabilistic Generative ModelsA deep dive into the architecture that taught neural networks to reason about uncertainty — covering autoencoders, the VAE architecture, μ…
Arko Sengupta·Oct 19Variational Autoencoder: A Comprehensive GuideExplore detailed Variational AutoencoderA response icon1A response icon1
prabath weerasinghe·Sep 21Demystifying the Variational AutoEncoderThe main text tells the story, and I wanted to keep it with as few equations as possible since the concepts are quite beautiful and…
Zeromathai·Sep 16Deep Generative Models: Four Ways to Make Complex Distributions LearnableAutoregressive models, VAEs, normalizing flows, and GANs address the same distribution-learning problem through fundamentally different stru
InFandhebyNancy·Sep 11The VAE Lives On as a Component, Not a GeneratorStable Diffusion’s “VAE” and Kingma & Welling’s VAE share a name and a shape, but not a training recipe. Reconstruction FID: 2.41 to 1.06.
Dhairya Thareja·Sep 4What are Autoencoders and Variational Autoencoders?Imagine that you have a large suitcase but very little space. You must pack only what matters. Later, you should be able to open the…
Karthikanivas·Aug 24What is an Autoencoder?An autoencoder is a type of neural network that learns to compress data into a smaller representation and then reconstruct the original…
InThe Pipeline: AI ML and CloudbyMichaelT Shomsky·Jul 27Automatic Placement of Tetrominos, Using a Transformer Model to Inform Best Placement by AgentsTraining a discrete n-by-n puzzle to assemble itself.
Pankaj Kumar·Jul 18Mapping Market States with VAEUsing latent space to uncover trend strength, volatility stress, mean-reversion pressure, and macro commodity context in market dataA response icon1A response icon1
