A two-dimensional analytical potential energy surface introduced in 1979 for testing optimization algorithms. It features three minima and curved transition pathways that evaluate an algorithm’s ability to navigate non-trivial topologies.
Step-by-step implementation of the classic Müller-Brown potential in PyTorch, with performance comparisons between analytical and automatic differentiation approaches for molecular dynamics and machine learning applications.
A high-performance, GPU-accelerated PyTorch testbed for ML-MD algorithms featuring JIT-compiled analytical Jacobian force kernels achieving 3-10x speedup over autograd, robust Langevin dynamics with Velocity-Verlet integration, and modular architecture designed as ground-truth validation for novel machine learning approaches in molecular dynamics.
ICML 2025 methodological paper introducing SPHNet, which uses adaptive network sparsification to overcome the computational bottleneck of tensor products in SE(3)-equivariant networks, achieving up to 7x speedup and 75% memory reduction on DFT Hamiltonian prediction tasks.
A NeurIPS 2021 method paper introducing Noise Contrastive Priors to address the VAE ‘prior hole’ problem, where standard Gaussian priors assign high density to regions of latent space that don’t correspond to realistic data, using energy-based models trained with contrastive learning to match the aggregate posterior.
A ground-up PyTorch implementation of Word2Vec treating it as a systems engineering challenge, with “tensorized tree” architecture converting pointer-chasing Hierarchical Softmax into dense GPU operations, infinite streaming datasets with Zipfian subsampling, and torch.compile compatibility for production-grade efficiency.
A complete guide to implementing modern Variational Autoencoders in PyTorch. Includes a copy-pasteable implementation, explanation of KL annealing to fix posterior collapse, and a deep dive into stable standard deviation parameterizations.
Learn to align molecular structures and point clouds using the Kabsch algorithm, with differentiable implementations for modern ML frameworks.
A PyTorch implementation enforcing strict Lyapunov stability guarantees on recurrent neural network controllers through Integral Quadratic Constraints, bridging 1990s robust control theory with modern deep reinforcement learning by solving semidefinite programs inside the gradient descent loop to provide mathematical certificates of safety.
We develop EigenNoise, a zero-data initialization method for word vectors that synthesizes representations from Zipf’s Law alone, demonstrating competitive performance to GloVe after fine-tuning without requiring any pre-training corpus.
Key dimensions that have helped me understand multi-arm bandit problems: action space, problem structure, external information, reward mechanism, and learner feedback.
Discover how NEAT and HyperNEAT changed neuroevolution by automatically designing neural network architectures and scaling them through geometric patterns.