NeurIPS 2024 accepted papers on Probabilistic and scientific modeling
390 of the 4,560 papers accepted at NeurIPS 2024 (8.6%) are on Probabilistic and scientific modeling, compared with 10.6% at NeurIPS 2023. They fall into 8 topics.
See NeurIPS 2024 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 · 99 papers
- Lorentz-Equivariant Geometric Algebra Transformers for High-Energy Physics (14 citations)
- Approximating mutual information of high-dimensional variables using learned representations (11 citations)
- Transferable Boltzmann Generators (11 citations)
- RGFN: Synthesizable Molecular Generation Using GFlowNets (10 citations)
- ReactZyme: A Benchmark for Enzyme-Reaction Prediction (7 citations)
- A benchmark for prediction of transcriptomic responses to chemical perturbations across cell types (6 citations)
- Designing Cell-Type-Specific Promoter Sequences Using Conservative Model-Based Optimization (6 citations)
- Optimal ablation for interpretability (6 citations)
- Aligning Target-Aware Molecule Diffusion Models with Exact Energy Optimization (5 citations)
- FlexSBDD: Structure-Based Drug Design with Flexible Protein Modeling (5 citations)
- Sequence-Augmented SE(3)-Flow Matching For Conditional Protein Generation (5 citations)
- A Generative Model of Symmetry Transformations (4 citations)
and 87 more in the interactive map.
Variational and generative sampling · 66 papers
- Discrete Flow Matching (16 citations)
- Optimal Flow Matching: Learning Straight Trajectories in Just One Step (11 citations)
- Principled Probabilistic Imaging using Diffusion Models as Plug-and-Play Priors (11 citations)
- Estimating the Hallucination Rate of Generative AI (9 citations)
- QVAE-Mole: The Quantum VAE with Spherical Latent Variable Learning for 3-D Molecule Generation (9 citations)
- Fisher Flow Matching for Generative Modeling over Discrete Data (6 citations)
- Categorical Flow Matching on Statistical Manifolds (5 citations)
- Latent Intrinsics Emerge from Training to Relight (5 citations)
- Poisson Variational Autoencoder (5 citations)
- Mutual Information Estimation via $f$-Divergence and Data Derangements (4 citations)
- Unity by Diversity: Improved Representation Learning for Multimodal VAEs (4 citations)
- Variational Flow Matching for Graph Generation (4 citations)
and 54 more in the interactive map.
Dynamical systems · 57 papers
- Universal Physics Transformers: A Framework For Efficiently Scaling Neural Operators (23 citations)
- The Well: a Large-Scale Collection of Diverse Physics Simulations for Machine Learning (10 citations)
- NeuroBOLT: Resting-state EEG-to-fMRI Synthesis with Multi-dimensional Feature Mapping (9 citations)
- Time Makes Space: Emergence of Place Fields in Networks Encoding Temporally Continuous Sensory Experiences (8 citations)
- Partial observation can induce mechanistic mismatches in data-constrained models of neural dynamics (7 citations)
- Input-to-State Stable Coupled Oscillator Networks for Closed-form Model-based Control in Latent Space (6 citations)
- WindsorML: High-Fidelity Computational Fluid Dynamics Dataset For Automotive Aerodynamics (5 citations)
- Inferring stochastic low-rank recurrent neural networks from neural data (4 citations)
- MiSO: Optimizing brain stimulation to create neural activity states (4 citations)
- Modeling Latent Neural Dynamics with Gaussian Process Switching Linear Dynamical Systems (4 citations)
- WildPPG: A Real-World PPG Dataset of Long Continuous Recordings (4 citations)
- Back to the Continuous Attractor (3 citations)
and 45 more in the interactive map.
Causal discovery · 52 papers
- Causal-learn: Causal Discovery in Python (26 citations)
- Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data (12 citations)
- CausalStock: Deep End-to-end Causal Discovery for News-driven Multi-stock Movement Prediction (5 citations)
- From Causal to Concept-Based Representation Learning (5 citations)
- A hierarchical decomposition for explaining ML performance discrepancies (4 citations)
- Tangent Space Causal Inference: Leveraging Vector Fields for Causal Discovery in Dynamical Systems (4 citations)
- Causal Deciphering and Inpainting in Spatio-Temporal Dynamics via Diffusion Model (3 citations)
- Learning Linear Causal Representations from General Environments: Identifiability and Intrinsic Ambiguity (3 citations)
- Marrying Causal Representation Learning with Dynamical Systems for Science (3 citations)
- Causal Dependence Plots (2 citations)
- Detecting and Measuring Confounding Using Causal Mechanism Shifts (2 citations)
- Hybrid Top-Down Global Causal Discovery with Local Search for Linear and Nonlinear Additive Noise Models (2 citations)
and 40 more in the interactive map.
Uncertainty and Gaussian processes · 41 papers
- Estimating Epistemic and Aleatoric Uncertainty with a Single Model (17 citations)
- Benchmarking Uncertainty Disentanglement: Specialized Uncertainties for Specialized Tasks (10 citations)
- An Information Theoretic Perspective on Conformal Prediction (5 citations)
- Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage? (5 citations)
- Neural Conditional Probability for Uncertainty Quantification (5 citations)
- Hierarchical Uncertainty Exploration via Feedforward Posterior Trees (4 citations)
- Robust Gaussian Processes via Relevance Pursuit (4 citations)
- Active Sequential Posterior Estimation for Sample-Efficient Simulation-Based Inference (3 citations)
- Divide-and-Conquer Posterior Sampling for Denoising Diffusion priors (3 citations)
- Kermut: Composite kernel regression for protein variant effects (3 citations)
- Practical Bayesian Algorithm Execution via Posterior Sampling (2 citations)
- Self-Calibrating Conformal Prediction (2 citations)
and 29 more in the interactive map.
Differential equations and PDEs · 37 papers
- Fourier Neural Operator with Learned Deformations for PDEs on General Geometries (150 citations)
- PINNacle: A Comprehensive Benchmark of Physics-Informed Neural Networks for Solving PDEs (28 citations)
- DiffusionPDE: Generative PDE-Solving under Partial Observation (26 citations)
- Latent Neural Operator for Solving Forward and Inverse PDE Problems (14 citations)
- Poseidon: Efficient Foundation Models for PDEs (12 citations)
- Newton Informed Neural Operator for Solving Nonlinear Partial Differential Equations (9 citations)
- APEBench: A Benchmark for Autoregressive Neural Emulators of PDEs (8 citations)
- AdaFlow: Imitation Learning with Variance-Adaptive Flow-Based Policies (7 citations)
- RoPINN: Region Optimized Physics-Informed Neural Networks (6 citations)
- Amortized Fourier Neural Operators (5 citations)
- Derivative-enhanced Deep Operator Network (5 citations)
- Data-Efficient Operator Learning via Unsupervised Pretraining and In-Context Learning (4 citations)
and 25 more in the interactive map.
Treatment effects · 20 papers
- Addressing Hidden Confounding with Heterogeneous Observational Datasets for Recommendation (3 citations)
- A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding (1 citation)
- Causal Contrastive Learning for Counterfactual Regression Over Time (1 citation)
- DiffPO: A causal diffusion model for learning distributions of potential outcomes (1 citation)
- Estimating Heterogeneous Treatment Effects by Combining Weak Instruments and Observational Data (1 citation)
- Hierarchical and Density-based Causal Clustering (1 citation)
- Higher-Order Causal Message Passing for Experimentation with Complex Interference (1 citation)
- Proximal Causal Inference With Text Data (1 citation)
- Using Surrogates in Covariate-adjusted Response-adaptive Randomization Experiments with Delayed Outcomes (1 citation)
- Causal Inference in the Closed-Loop: Marginal Structural Models for Sequential Excursion Effects
- Conditional Outcome Equivalence: A Quantile Alternative to CATE
- Dynamic Subgroup Identification in Covariate-adjusted Response-adaptive Randomization Experiments
and 8 more in the interactive map.
Bayesian optimization · 18 papers
- Cost-aware Bayesian Optimization via the Pandora's Box Gittins Index (5 citations)
- A survey and benchmark of high-dimensional Bayesian optimization of discrete sequences (3 citations)
- Transition Constrained Bayesian Optimization via Markov Decision Processes (3 citations)
- Generative Adversarial Model-Based Optimization via Source Critic Regularization (2 citations)
- Active Learning for Derivative-Based Global Sensitivity Analysis with Gaussian Processes (1 citation)
- Batched Energy-Entropy acquisition for Bayesian Optimization (1 citation)
- Inversion-based Latent Bayesian Optimization (1 citation)
- Model Fusion through Bayesian Optimization in Language Model Fine-Tuning (1 citation)
- This Too Shall Pass: Removing Stale Observations in Dynamic Bayesian Optimization (1 citation)
- Active Set Ordering
- Approximation-Aware Bayesian Optimization
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal
and 6 more in the interactive map.