Dispatches from the edge of chaos — on nonlinear dynamics, AI, emergence, and the mathematics of complex systems.
ImageGPT (iGPT) applies the autoregressive GPT architecture directly to image generation by treating images as sequences of pixels or color clusters, demonstrating that language model approaches
The Variational Autoencoder (VAE) is a generative model that learns a continuous latent representation of data by combining an autoencoder architecture with variational Bayesian inference, enabling
StyleGAN is a revolutionary generative architecture that produces photorealistic images by borrowing from neural style transfer, using a mapping network and adaptive instance normalization to control
Generative Adversarial Networks (GANs) learn to generate realistic data through an adversarial game between two neural networks—a generator that creates samples and a discriminator that
Neural Ordinary Differential Equations (Neural ODEs) replace discrete layer-by-layer transformations with continuous dynamics defined by neural networks, treating depth as a continuous variable and computing outputs
Consistency Models are a new family of generative models that enable high-quality single-step image generation by learning to map any point along a diffusion trajectory directly
Flow Matching is a generative modeling framework that learns continuous normalizing flows by regressing onto simple vector fields, providing a simpler and more flexible alternative to
VQ-VAE (Vector Quantized Variational Autoencoder) and VQGAN (Vector Quantized GAN) learn discrete codebook representations of images, enabling powerful image generation by converting the continuous pixel space