ICLR 2023 accepted papers on Probabilistic and scientific modeling
148 of the 1,575 papers accepted at ICLR 2023 (9.4%) are on Probabilistic and scientific modeling, compared with 10.0% at ICLR 2022. They fall into 8 topics.
See ICLR 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Molecules and proteins · 41 papers
- DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking (362 citations)
- Diffusion Probabilistic Modeling of Protein Backbones in 3D for the motif-scaffolding problem (95 citations)
- PiFold: Toward effective and efficient protein inverse folding (53 citations)
- Pre-training via Denoising for Molecular Property Prediction (30 citations)
- Rotamer Density Estimator is an Unsupervised Learner of the Effect of Mutations on Protein-Protein Interaction (29 citations)
- A Graph Neural Network Approach to Automated Model Building in Cryo-EM Maps (27 citations)
- Conditional Antibody Design as 3D Equivariant Graph Translation (26 citations)
- Learning Hierarchical Protein Representations via Complete 3D Graph Networks (16 citations)
- E3Bind: An End-to-End Equivariant Network for Protein-Ligand Docking (15 citations)
- Equivariant Energy-Guided SDE for Inverse Molecular Design (15 citations)
- Symbolic Physics Learner: Discovering governing equations via Monte Carlo tree search (14 citations)
- Protein Sequence and Structure Co-Design with Equivariant Translation (13 citations)
and 29 more in the interactive map.
Variational and generative sampling · 36 papers
- Flow Matching for Generative Modeling (108 citations)
- Building Normalizing Flows with Stochastic Interpolants (23 citations)
- Flow Annealed Importance Sampling Bootstrap (14 citations)
- Explicitly Minimizing the Blur Error of Variational Autoencoders (8 citations)
- Backpropagation at the Infinitesimal Inference Limit of Energy-Based Models: Unifying Predictive Coding, Equilibrium Propagation, and Contrastive Hebbian Learning (6 citations)
- Classically Approximating Variational Quantum Machine Learning with Random Fourier Features (6 citations)
- Scaling Up Probabilistic Circuits by Latent Variable Distillation (5 citations)
- Guiding Energy-based Models via Contrastive Latent Variables (4 citations)
- Quantized Compressed Sensing with Score-Based Generative Models (4 citations)
- Statistical Efficiency of Score Matching: The View from Isoperimetry (4 citations)
- DAVA: Disentangling Adversarial Variational Autoencoder (3 citations)
- Multi-Rate VAE: Train Once, Get the Full Rate-Distortion Curve (3 citations)
and 24 more in the interactive map.
Uncertainty and Gaussian processes · 18 papers
- TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second (102 citations)
- Packed Ensembles for efficient uncertainty estimation (13 citations)
- Meta-learning Adaptive Deep Kernel Gaussian Processes for Molecular Property Prediction (9 citations)
- Autoregressive Conditional Neural Processes (3 citations)
- Gradient Boosting Performs Gaussian Process Inference (3 citations)
- A Laplace-inspired Distribution on SO(3) for Probabilistic Rotation Estimation (2 citations)
- Active Learning in Bayesian Neural Networks with Balanced Entropy Learning Principle (2 citations)
- MARS: Meta-learning as Score Matching in the Function Space (1 citation)
- Sampling-based inference for large linear models, with application to linearised Laplace (1 citation)
- Calibrating Transformers via Sparse Gaussian Processes
- Deep Variational Implicit Processes
- Martingale Posterior Neural Processes
and 6 more in the interactive map.
Dynamical systems · 18 papers
- Actionable Neural Representations: Grid Cells from Minimal Constraints (17 citations)
- Continuous PDE Dynamics Forecasting with Implicit Neural Representations (13 citations)
- Disentanglement with Biological Constraints: A Theory of Functional Cell Types (13 citations)
- FluidLab: A Differentiable Environment for Benchmarking Complex Fluid Manipulation (11 citations)
- Spatial Attention Kinetic Networks with E(n)-Equivariance (11 citations)
- A Theoretical Framework for Inference and Learning in Predictive Coding Networks (8 citations)
- Temporal Domain Generalization with Drift-Aware Dynamic Neural Networks (6 citations)
- Understanding Neural Coding on Latent Manifolds by Sharing Features and Dividing Ensembles (5 citations)
- Evolve Smoothly, Fit Consistently: Learning Smooth Latent Dynamics For Advection-Dominated Systems (4 citations)
- A probabilistic framework for task-aligned intra- and inter-area neural manifold estimation (3 citations)
- Latent Neural ODEs with Sparse Bayesian Multiple Shooting (3 citations)
- Neural Lagrangian Schr\"{o}dinger Bridge: Diffusion Modeling for Population Dynamics (1 citation)
and 6 more in the interactive map.
Differential equations and PDEs · 15 papers
- Clifford Neural Layers for PDE Modeling (20 citations)
- Competitive Physics Informed Networks (10 citations)
- CROM: Continuous Reduced-Order Modeling of PDEs Using Implicit Neural Representations (10 citations)
- Factorized Fourier Neural Operators (9 citations)
- Learning differentiable solvers for systems with hard constraints (6 citations)
- FINDE: Neural Differential Equations for Finding and Preserving Invariant Quantities (4 citations)
- Learning Controllable Adaptive Simulation for Multi-resolution Physics (4 citations)
- Guiding continuous operator learning through Physics-based boundary constraints (3 citations)
- NTFields: Neural Time Fields for Physics-Informed Robot Motion Planning (3 citations)
- Bi-level Physics-Informed Neural Networks for PDE Constrained Optimization using Broyden's Hypergradients (2 citations)
- Characteristic Neural Ordinary Differential Equation (2 citations)
- A Stable and Scalable Method for Solving Initial Value PDEs with Neural Networks (1 citation)
and 3 more in the interactive map.
Causal discovery · 13 papers
- CUTS: Neural Causal Discovery from Irregular Time-Series Data (9 citations)
- Predicting Cellular Responses with Variational Causal Inference and Refined Relational Information (6 citations)
- Can We Faithfully Represent Absence States to Compute Shapley Values on a DNN? (5 citations)
- Diffusion Models for Causal Discovery via Topological Ordering (4 citations)
- Rhino: Deep Causal Temporal Relationship Learning with History-dependent Noise (4 citations)
- Causal Reasoning in the Presence of Latent Confounders via Neural ADMG Learning (3 citations)
- Neural Causal Models for Counterfactual Identification and Estimation (3 citations)
- Measuring axiomatic soundness of counterfactual image models (2 citations)
- Causal Representation Learning for Instantaneous and Temporal Effects in Interactive Systems
- GRACE-C: Generalized Rate Agnostic Causal Estimation via Constraints
- Tier Balancing: Towards Dynamic Fairness over Underlying Causal Factors
- Boosting Causal Discovery via Adaptive Sample Reweighting
and 1 more in the interactive map.