Importance Weighted Autoencoders: Beyond the Standard VAE
The key difference between multi-sample VAEs and IWAEs: how log-of-averages creates a tighter bound on log-likelihood.
Optimization - Page 2
IWAE: Importance Weighted Autoencoders
Summary of Burda, Grosse & Salakhutdinov's ICLR 2016 paper introducing Importance Weighted Autoencoders for tighter …
How to Fold Graciously: The Levinthal Paradox
A perspective paper defining the Grand Challenge of protein folding: distinguishing kinetic pathways from thermodynamic …
Müller-Brown Potential
The Müller-Brown potential is a classic 2D benchmark for testing optimization algorithms and molecular dynamics methods.
Implementing the Müller-Brown Potential in PyTorch
Guide to implementing the Müller-Brown potential in PyTorch, comparing analytical vs automatic differentiation with …
Müller-Brown Potential: A PyTorch ML Testbed
GPU-accelerated PyTorch framework for the Müller-Brown potential with JIT compilation, Langevin dynamics, and …
Efficient DFT Hamiltonian Prediction via Adaptive Sparsity
Luo et al. introduce SPHNet, using adaptive sparsity to dramatically improve SE(3)-equivariant Hamiltonian prediction …
Contrastive Learning for Variational Autoencoder Priors
Aneja et al.'s NeurIPS 2021 paper introducing Noise Contrastive Priors (NCPs) to address VAE's 'prior hole' problem with …
High-Performance Word2Vec in Pure PyTorch
Production-grade Word2Vec in PyTorch with vectorized Hierarchical Softmax, Negative Sampling, and torch.compile support.
Modern PyTorch VAEs: A Detailed Implementation Guide
Complete PyTorch VAE tutorial: Copy-paste code, ELBO derivation, KL annealing, and stable softplus parameterization.
IQCRNN: Certified Stability for Neural Networks
PyTorch IQCRNN enforcing stability guarantees on RNNs via Integral Quadratic Constraints and semidefinite programming.
EigenNoise: Data-Free Word Vector Initialization
Investigation into EigenNoise, a data-free initialization scheme for word vectors that approaches pre-trained model …