NeurIPS 2023 accepted papers on Probabilistic and scientific modeling
380 of the 3,584 papers accepted at NeurIPS 2023 (10.6%) are on Probabilistic and scientific modeling, compared with 10.4% at NeurIPS 2022. They fall into 8 topics.
See NeurIPS 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 · 93 papers
- ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design (30 citations)
- PoET: A generative model of protein families as sequences-of-sequences (29 citations)
- PreDiff: Precipitation Nowcasting with Latent Diffusion Models (29 citations)
- Tartarus: A Benchmarking Platform for Realistic And Practical Inverse Molecular Design (28 citations)
- ProteinNPT: Improving Protein Property Prediction and Design with Non-Parametric Transformers (24 citations)
- DiffPack: A Torsional Diffusion Model for Autoregressive Protein Side-Chain Packing (16 citations)
- Predicting a Protein's Stability under a Million Mutations (16 citations)
- Transformer-based Planning for Symbolic Regression (16 citations)
- OpenProteinSet: Training data for structural biology at scale (15 citations)
- Prefix-Tree Decoding for Predicting Mass Spectra from Molecules (13 citations)
- Functional-Group-Based Diffusion for Pocket-Specific Molecule Generation and Elaboration (11 citations)
- Predicting mutational effects on protein-protein binding via a side-chain diffusion probabilistic model (11 citations)
and 81 more in the interactive map.
Variational and generative sampling · 74 papers
- Beyond Normal: On the Evaluation of Mutual Information Estimators (15 citations)
- Equivariant flow matching (12 citations)
- Flow Matching for Scalable Simulation-Based Inference (8 citations)
- Distributional Learning of Variational AutoEncoder: Application to Synthetic Data Generation (6 citations)
- Gacs-Korner Common Information Variational Autoencoder (6 citations)
- Max-Sliced Mutual Information (6 citations)
- Energy-Based Sliced Wasserstein Distance (5 citations)
- Hyperbolic VAE via Latent Gaussian Distributions (5 citations)
- Bridging Discrete and Backpropagation: Straight-Through and Beyond (4 citations)
- Canonical normalizing flows for manifold learning (4 citations)
- Joint Bayesian Inference of Graphical Structure and Parameters with a Single Generative Flow Network (4 citations)
- Adapting to Continuous Covariate Shift via Online Density Ratio Estimation (3 citations)
and 62 more in the interactive map.
Uncertainty and Gaussian processes · 62 papers
- GAUCHE: A Library for Gaussian Processes in Chemistry (33 citations)
- Pretraining task diversity and the emergence of non-Bayesian in-context learning for regression (11 citations)
- Neural Circuits for Fast Poisson Compressed Sensing in the Olfactory Bulb (10 citations)
- Learning Robust Statistics for Simulation-based Inference under Model Misspecification (9 citations)
- Topology-Aware Uncertainty for Image Segmentation (9 citations)
- Effective Bayesian Heteroscedastic Regression with Deep Neural Networks (7 citations)
- Variational Imbalanced Regression: Fair Uncertainty Quantification via Probabilistic Smoothing (7 citations)
- Compression with Bayesian Implicit Neural Representations (6 citations)
- Quantification of Uncertainty with Adversarial Models (5 citations)
- Beyond Deep Ensembles: A Large-Scale Evaluation of Bayesian Deep Learning under Distribution Shift (4 citations)
- A Bayesian Approach To Analysing Training Data Attribution In Deep Learning (3 citations)
- Bayesian target optimisation for high-precision holographic optogenetics (3 citations)
and 50 more in the interactive map.
Causal discovery · 47 papers
- Identifiability Guarantees for Causal Disentanglement from Soft Interventions (9 citations)
- Unpaired Multi-Domain Causal Representation Learning (9 citations)
- Nonparametric Identifiability of Causal Representations from Unknown Interventions (7 citations)
- BayesDAG: Gradient-Based Posterior Inference for Causal Discovery (6 citations)
- Causal normalizing flows: from theory to practice (5 citations)
- Comparing Causal Frameworks: Potential Outcomes, Structural Models, Graphs, and Abstractions (5 citations)
- Zero-shot causal learning (5 citations)
- Causal de Finetti: On the Identification of Invariant Causal Structure in Exchangeable Data (4 citations)
- Learning Linear Causal Representations from Interventions under General Nonlinear Mixing (4 citations)
- Causal discovery from observational and interventional data across multiple environments (3 citations)
- DynGFN: Towards Bayesian Inference of Gene Regulatory Networks with GFlowNets (3 citations)
- Characterization and Learning of Causal Graphs with Small Conditioning Sets (2 citations)
and 35 more in the interactive map.
Dynamical systems · 44 papers
- Cognitive Model Discovery via Disentangled RNNs (27 citations)
- DYffusion: A Dynamics-informed Diffusion Model for Spatiotemporal Forecasting (21 citations)
- Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamics (20 citations)
- An information-theoretic quantification of the content of communication between brain regions (19 citations)
- Score-based Data Assimilation (13 citations)
- Implicit Transfer Operator Learning: Multiple Time-Resolution Models for Molecular Dynamics (7 citations)
- Hierarchical VAEs provide a normative account of motion processing in the primate brain (6 citations)
- Information Geometry of the Retinal Representation Manifold (6 citations)
- Neural Oscillators are Universal (4 citations)
- Bypassing spike sorting: Density-based decoding using spike localization from dense multielectrode probes (3 citations)
- Gaussian Partial Information Decomposition: Bias Correction and Application to High-dimensional Data (3 citations)
- Neural Foundations of Mental Simulation: Future Prediction of Latent Representations on Dynamic Scenes (3 citations)
and 32 more in the interactive map.
Differential equations and PDEs · 25 papers
- PDE-Refiner: Achieving Accurate Long Rollouts with Neural PDE Solvers (23 citations)
- Operator Learning with Neural Fields: Tackling PDEs on General Geometries (15 citations)
- Generalization bounds for neural ordinary differential equations and deep residual networks (8 citations)
- Hypernetwork-based Meta-Learning for Low-Rank Physics-Informed Neural Networks (7 citations)
- Domain Agnostic Fourier Neural Operators (5 citations)
- Lie Point Symmetry and Physics-Informed Networks (5 citations)
- Self-Consistent Velocity Matching of Probability Flows (5 citations)
- StEik: Stabilizing the Optimization of Neural Signed Distance Functions and Finer Shape Representation (5 citations)
- Deep Equilibrium Based Neural Operators for Steady-State PDEs (3 citations)
- Entropy-dissipation Informed Neural Network for McKean-Vlasov Type PDEs (3 citations)
- Neural Ideal Large Eddy Simulation: Modeling Turbulence with Neural Stochastic Differential Equations (3 citations)
- Amortized Reparametrization: Efficient and Scalable Variational Inference for Latent SDEs (2 citations)
and 13 more in the interactive map.
Bayesian optimization · 19 papers
- Unexpected Improvements to Expected Improvement for Bayesian Optimization (48 citations)
- Framework and Benchmarks for Combinatorial and Mixed-variable Bayesian Optimization (6 citations)
- Sample-efficient Multi-objective Molecular Optimization with GFlowNets (6 citations)
- Human-in-the-Loop Optimization for Deep Stimulus Encoding in Visual Prostheses (5 citations)
- Bounce: Reliable High-Dimensional Bayesian Optimization for Combinatorial and Mixed Spaces (4 citations)
- PROTES: Probabilistic Optimization with Tensor Sampling (3 citations)
- Scaling Laws for Hyperparameter Optimization (3 citations)
- Quantum Bayesian Optimization (2 citations)
- Self-Correcting Bayesian Optimization through Bayesian Active Learning (2 citations)
- Distributionally Robust Bayesian Optimization with $\varphi$-divergences (1 citation)
- Efficient Robust Bayesian Optimization for Arbitrary Uncertain inputs (1 citation)
- Neural Latent Geometry Search: Product Manifold Inference via Gromov-Hausdorff-Informed Bayesian Optimization (1 citation)
and 7 more in the interactive map.
Treatment effects · 16 papers
- Optimal Transport for Treatment Effect Estimation (7 citations)
- Double and Single Descent in Causal Inference with an Application to High-Dimensional Synthetic Control (6 citations)
- Sharp Bounds for Generalized Causal Sensitivity Analysis (2 citations)
- A Cross-Moment Approach for Causal Effect Estimation (1 citation)
- High Precision Causal Model Evaluation with Conditional Randomization (1 citation)
- Marginal Density Ratio for Off-Policy Evaluation in Contextual Bandits (1 citation)
- Optimal Treatment Regimes for Proximal Causal Learning (1 citation)
- Optimized Covariance Design for AB Test on Social Network under Interference (1 citation)
- Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations (1 citation)
- Temporal Causal Mediation through a Point Process: Direct and Indirect Effects of Healthcare Interventions (1 citation)
- An Efficient Doubly-Robust Test for the Kernel Treatment Effect
- Estimating Propensity for Causality-based Recommendation without Exposure Data
and 4 more in the interactive map.