The purpose of this note is to act as a repository for all the literature read during my PhD. The literature read has been segregated in different domains ranging from dynamics to representing molecule as a 3d object.
- Paper Count = 77 /1280
+ Reading = 5-
Dynamics
- ODE2VAE
- Neural ODEs
- Neural Flows
- FFJORD
- Neural Processes
- Review of Normalizing flows
- Variational multiple shooting for Bayesian ODEs with Gaussian processes
- Non-Autoregressive neural machine translation
- Your classifier is secretly an energy based model and you should treat it like one
- Neural Relational Inference for Interacting systems
- Learning Continuous-time PDEs from sparse data with GNNs
- PointFlow
- Conditional Random fields
- Message Passing Neural PDE Solvers
- Neural Controlled Differential Equations
- Variational Neural Cellular Automata
- Learning to generate 3D shapes with generative cellular automata
- Neural process with stochastic attention
-
Graph based learning
- Graph Condensation for Graph Neural Networks
- Simple GNN Regularisation for 3D Molecule Property prediction
- Graph Normalizing flows
- Evolving-Graph Gaussian Processes
- Handling distribution shifts in graphs
- Learning to Solve PDE-constrained Inverse Problems with Graph Networks
- Topological graph neural networks
- PDE-GCN: Novel Architectures for Graph Neural Networks Motivated by Partial Differential Equations
- Spherical Message passing for 3D Graph Networks
- Temporal Graph Networks
- E(n) Equivariant Graph Neural Networks
- Graph Coupled Oscillator Networks
-
Physics Inspired ML
- Noether Networks
- Interaction Networks for Learning about Objects,Relations and Physics
- Physics Informed Machine Learning
- Deconstructing the inductive biases of Hamiltonian neural networks
- Hamiltonian graph networks with ODE integrators
- Predicting physics in mesh-reduced space with temporal attention
-
Conformer generation and structure prediction
- Learning Gradient Fields for Molecular Conformation Generation
- DGSM
- Learning neural generative dynamics for molecular conformation generation
- Geometric Deep Learning on Molecular Representations
- EQUIBIND: Geometric Deep Learning for Drug Binding Structure Prediction
- Spanning Tree-based Graph Generation for Molecules
- Molecular Surface Representation Using 3D Zernike Descriptors for Protein Shape Comparison and Docking
- Categorical Normalizing Flow
- GraphDF: A Discrete Flow Model for Molecular Graph Generation
- GraphNVP
- GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles
- Dual use of artifcial-intelligence-powered drug discovery
- Energy-Inspired Molecular Conformation Optimization
- Chemical-Reaction-Aware Molecule Representation Learning
- Evaluating generalization in Gflow Nets for molecule design
- An auto regressive flow model for 3D molecular geometry generation from scratch
- Molecular RNN
- Learning to extend molecular scaffolds with structural motifs
- Learning 3D representations of molecular chirality with invariance to bond rotations
- Pre-training molecular graph representation with 3D geometry
- DATA-EFFICIENT GRAPH GRAMMAR LEARNING FOR MOLECULAR GENERATION
- Geometric Transformers for protein interface contact prediction
- A 3D Molecule Generative Model for Structure-Based Drug Design
- Crystal Diffusion Variational Autoencoder for Periodic Material Generation
- Generative Coarse-Graining of Molecular Conformations
- Generating 3D Molecules for Target Protein Binding
-
Diffusion, SDE and Score Matching Methods
- GeoDiff
- Diffusion Kernels on Graphs and Other Discrete Input Spaces
- Denoising Probabilistic Diffusion models
- Estimation of Non-Normalized Statistical Models by Score Matching
- GRAND: Graph Neural Diffusion
- Learning energy-based models by diffusion recovery likelihood
- Denoising Diffusion GANs
- Scalable Gradients for SDE
- Graph Anisotropic Diffusion
- Auto-Regressive Diffusion Models
- GRAND++
- Reimaninan Neural SDE
- Neural Sheaf Diffusion
- Neural SDEs as Infinite-Dimensional GANs
- Score-based generative modeling through SDE
-
RL