Search Results for author:
Found 13 papers, 5 papers with code
Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces
Generalizing from specific (Gaussian) forward processes to discrete-state processes without a variational approximation sheds light on how to interpret diffusion models, which we discuss.
Ranked #13 on
Image Generation
on CelebA 64x64
Predictive Scale-Bridging Simulations through Active Learning
Throughout computational science, there is a growing need to utilize the continual improvements in raw computational horsepower to achieve greater physical fidelity through scale-bridging over brute-force increases in the number of mesh elements.
Multi-Scale Neural Networks for to Fluid Flow in 3D Porous Media
The permeability of complex porous materials can be obtained via direct flow simulation, which provides the most accurate results, but is very computationally expensive.
Simple and efficient algorithms for training machine learning potentials to force data
These forces provide much more information than the energy alone.
Modeling nanoconfinement effects using active learning
At this scale, the fluid properties are affected by nanoconfinement effects due to the increased fluid-solid interactions.
Automated discovery of a robust interatomic potential for aluminum
The accuracy and robustness of an ML potential is primarily limited by the quality and diversity of the training dataset.
Embedding Hard Physical Constraints in Convolutional Neural Networks for 3D Turbulence
Deep learning approaches have shown much promise for physical sciences, especially in dimensionality reduction and compression of large datasets.
Embedding Hard Physical Constraints in Neural Network Coarse-Graining of 3D Turbulence
In the recent years, deep learning approaches have shown much promise in modeling complex systems in the physical sciences.
Computational Physics
Less is more: sampling chemical space with active learning
In this work, we present a fully automated approach for the generation of datasets with the intent of training universal ML potentials.
Hierarchical modeling of molecular energies using a deep neural network
We introduce the Hierarchically Interacting Particle Neural Network (HIP-NN) to model molecular properties from datasets of quantum calculations.
Ranked #13 on
Formation Energy
on QM9
Machine Learning Predicts Laboratory Earthquakes
Forecasting fault failure is a fundamental but elusive goal in earthquake science.
Geophysics
Inferring low-dimensional microstructure representations using convolutional neural networks
We use activations in a pre-trained convolutional neural network to provide a high-dimensional characterization of a set of synthetic microstructural images.
The Effective Field Theory of Dark Matter Direct Detection
We extend and explore the general non-relativistic effective theory of dark matter (DM) direct detection.
High Energy Physics - Phenomenology Cosmology and Nongalactic Astrophysics
