Efficient and Scalable Density Functional Theory Hamiltonian Prediction through Adaptive Sparsity
Luo et al. introduce SPHNet, using adaptive sparsity to dramatically improve SE(3)-equivariant Hamiltonian prediction …...
Neural-Networks
Learning Smooth and Expressive Interatomic Potentials for Physical Property Prediction
Fu et al. propose energy conservation as a key MLIP diagnostic and introduce eSEN, bridging test accuracy and real …...
The Dark Side of the Forces: Assessing Non-Conservative Force Models for Atomistic Machine Learning
Bigi et al. critique non-conservative force models in ML potentials, showing their simulation failures and proposing …...
A Contrastive Learning Approach for Training Variational Autoencoder Priors
Aneja et al.'s NeurIPS 2021 paper introducing Noise Contrastive Priors (NCPs) to address VAE's 'prior hole' problem with …...
3D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Data
Weiler et al.'s NeurIPS 2018 paper introducing SE(3)-equivariant CNNs for volumetric data using group theory and …...
Invalid SMILES are Beneficial Rather than Detrimental to Chemical Language Models
Skinnider (2024) shows that generating invalid SMILES actually improves chemical language model performance through …...
Modern PyTorch Techniques for VAEs: A Hands-On Tutorial
Learn to implement VAEs in PyTorch: ELBO objective, reparameterization trick, loss scaling, and MNIST experiments on …
IQCRNN: Certified Stability for Neural Networks
PyTorch IQCRNN enforcing stability guarantees on RNNs via Integral Quadratic Constraints and semidefinite programming....
IMG2SMI: Translating Molecular Structure Images to SMILES
Campos & Ji's method for converting 2D molecular images to SMILES strings using Transformers and SELFIES representation....
GPT-2 Susceptibility to Universal Adversarial Triggers
Investigation into whether universal adversarial triggers can control both topic and stance of GPT-2's generated text …...
HyperNEAT: Scaling Neuroevolution with Geometric Patterns
How HyperNEAT uses indirect encoding and geometric patterns to evolve large-scale neural networks with biological …
NEAT: Evolving Neural Network Topologies
Learn about NEAT's approach to evolving neural networks: automatic topology design, historical markings, and speciation …