NeurIPS 2022 accepted papers on Probabilistic and scientific modeling
302 of the 2,901 papers accepted at NeurIPS 2022 (10.4%) are on Probabilistic and scientific modeling, compared with 12.1% at NeurIPS 2021. They fall into 8 topics.
See NeurIPS 2022 in the interactive map
Submitting to NeurIPS? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Variational and generative sampling · 78 papers
- Manifold Interpolating Optimal-Transport Flows for Trajectory Inference (25 citations)
- Escaping from the Barren Plateau via Gaussian Initializations in Deep Variational Quantum Circuits (22 citations)
- Sampling with Riemannian Hamiltonian Monte Carlo in a Constrained Space (21 citations)
- DIMES: A Differentiable Meta Solver for Combinatorial Optimization Problems (16 citations)
- Variational inference via Wasserstein gradient flows (12 citations)
- Automatic Differentiation of Programs with Discrete Randomness (11 citations)
- Poisson Flow Generative Models (11 citations)
- A Geometric Perspective on Variational Autoencoders (10 citations)
- Wavelet Score-Based Generative Modeling (10 citations)
- Exploring the Latent Space of Autoencoders with Interventional Assays (9 citations)
- Learning Manifold Dimensions with Conditional Variational Autoencoders (9 citations)
- Weakly Supervised Representation Learning with Sparse Perturbations (9 citations)
and 66 more in the interactive map.
Uncertainty and Gaussian processes · 48 papers
- CARD: Classification and Regression Diffusion Models (33 citations)
- Truncated proposals for scalable and hassle-free simulation-based inference (25 citations)
- Training Uncertainty-Aware Classifiers with Conformalized Deep Learning (17 citations)
- Invariance Learning in Deep Neural Networks with Differentiable Laplace Approximations (11 citations)
- Pitfalls of Epistemic Uncertainty Quantification through Loss Minimisation (10 citations)
- Towards Reliable Simulation-Based Inference with Balanced Neural Ratio Estimation (10 citations)
- Bayesian Active Learning with Fully Bayesian Gaussian Processes (9 citations)
- On Uncertainty, Tempering, and Data Augmentation in Bayesian Classification (9 citations)
- Robust Neural Posterior Estimation and Statistical Model Criticism (9 citations)
- Single Model Uncertainty Estimation via Stochastic Data Centering (9 citations)
- Data Augmentation MCMC for Bayesian Inference from Privatized Data (8 citations)
- Nonparametric Uncertainty Quantification for Single Deterministic Neural Network (8 citations)
and 36 more in the interactive map.
Molecules and proteins · 43 papers
- TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction (171 citations)
- Motion Transformer with Global Intention Localization and Local Movement Refinement (73 citations)
- LasUIE: Unifying Information Extraction with Latent Adaptive Structure-aware Generative Language Model (69 citations)
- Predicting Cellular Responses to Novel Drug Perturbations at a Single-Cell Resolution (44 citations)
- Reinforced Genetic Algorithm for Structure-based Drug Design (28 citations)
- Non-identifiability and the Blessings of Misspecification in Models of Molecular Fitness (19 citations)
- Concept Activation Regions: A Generalized Framework For Concept-Based Explanations (16 citations)
- Semantic Probabilistic Layers for Neuro-Symbolic Learning (15 citations)
- So3krates: Equivariant attention for interactions on arbitrary length-scales in molecular systems (15 citations)
- DDXPlus: A New Dataset For Automatic Medical Diagnosis (12 citations)
- ENS-10: A Dataset For Post-Processing Ensemble Weather Forecasts (8 citations)
- Simultaneous Missing Value Imputation and Structure Learning with Groups (8 citations)
and 31 more in the interactive map.
Dynamical systems · 42 papers
- No Free Lunch from Deep Learning in Neuroscience: A Case Study through Models of the Entorhinal-Hippocampal Circuit (56 citations)
- Neural Lyapunov Control of Unknown Nonlinear Systems with Stability Guarantees (16 citations)
- Efficient coding, channel capacity, and the emergence of retinal mosaics (15 citations)
- The least-control principle for local learning at equilibrium (15 citations)
- Extracting computational mechanisms from neural data using low-rank RNNs (12 citations)
- Guaranteed Conservation of Momentum for Learning Particle-based Fluid Dynamics (12 citations)
- Distinguishing discrete and continuous behavioral variability using warped autoregressive HMMs (10 citations)
- Holomorphic Equilibrium Propagation Computes Exact Gradients Through Finite Size Oscillations (7 citations)
- Sparsity in Continuous-Depth Neural Networks (6 citations)
- Biologically-plausible backpropagation through arbitrary timespans via local neuromodulators (5 citations)
- Iterative Structural Inference of Directed Graphs (5 citations)
- The computational and learning benefits of Daleian neural networks (5 citations)
and 30 more in the interactive map.
Causal discovery · 31 papers
- Root Cause Analysis of Failures in Microservices through Causal Discovery (33 citations)
- Weakly supervised causal representation learning (13 citations)
- Latent Hierarchical Causal Structure Discovery with Rank Constraints (12 citations)
- MissDAG: Causal Discovery in the Presence of Missing Data with Continuous Additive Noise Models (11 citations)
- Sound and Complete Causal Identification with Latent Variables Given Local Background Knowledge (10 citations)
- Large-Scale Differentiable Causal Discovery of Factor Graphs (9 citations)
- Temporally Disentangled Representation Learning (8 citations)
- DeepMed: Semiparametric Causal Mediation Analysis with Debiased Deep Learning (7 citations)
- Amortized Inference for Causal Structure Learning (6 citations)
- CEBaB: Estimating the Causal Effects of Real-World Concepts on NLP Model Behavior (6 citations)
- Debiased, Longitudinal and Coordinated Drug Recommendation through Multi-Visit Clinic Records (6 citations)
- Active Bayesian Causal Inference (5 citations)
and 19 more in the interactive map.
Differential equations and PDEs · 22 papers
- DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps (284 citations)
- PDEBench: An Extensive Benchmark for Scientific Machine Learning (56 citations)
- AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier–Stokes Solutions (37 citations)
- Generic bounds on the approximation error for physics-informed (and) operator learning (26 citations)
- Is $L^2$ Physics Informed Loss Always Suitable for Training Physics Informed Neural Network? (24 citations)
- Accelerated Training of Physics-Informed Neural Networks (PINNs) using Meshless Discretizations (22 citations)
- Meta-Auto-Decoder for Solving Parametric Partial Differential Equations (16 citations)
- Learning to Accelerate Partial Differential Equations via Latent Global Evolution (14 citations)
- NeurOLight: A Physics-Agnostic Neural Operator Enabling Parametric Photonic Device Simulation (12 citations)
- Neural Conservation Laws: A Divergence-Free Perspective (10 citations)
- Neural Stochastic PDEs: Resolution-Invariant Learning of Continuous Spatiotemporal Dynamics (10 citations)
- M2N: Mesh Movement Networks for PDE Solvers (8 citations)
and 10 more in the interactive map.
Bayesian optimization · 21 papers
- Joint Entropy Search for Multi-Objective Bayesian Optimization (13 citations)
- Batch Bayesian Optimization on Permutations using the Acquisition Weighted Kernel (11 citations)
- Local Latent Space Bayesian Optimization over Structured Inputs (11 citations)
- Increasing the Scope as You Learn: Adaptive Bayesian Optimization in Nested Subspaces (10 citations)
- Monte Carlo Tree Search based Variable Selection for High Dimensional Bayesian Optimization (10 citations)
- Modeling the Machine Learning Multiverse (9 citations)
- Joint Entropy Search For Maximally-Informed Bayesian Optimization (8 citations)
- SnAKe: Bayesian Optimization with Pathwise Exploration (8 citations)
- Bidirectional Learning for Offline Infinite-width Model-based Optimization (7 citations)
- Tree ensemble kernels for Bayesian optimization with known constraints over mixed-feature spaces (5 citations)
- Triangulation candidates for Bayesian optimization (5 citations)
- Bandit Theory and Thompson Sampling-Guided Directed Evolution for Sequence Optimization (4 citations)
and 9 more in the interactive map.
Treatment effects · 17 papers
- Using Embeddings for Causal Estimation of Peer Influence in Social Networks (9 citations)
- Benchmarking Heterogeneous Treatment Effect Models through the Lens of Interpretability (6 citations)
- Counterfactual Neural Temporal Point Process for Estimating Causal Influence of Misinformation on Social Media (6 citations)
- Staggered Rollout Designs Enable Causal Inference Under Interference Without Network Knowledge (6 citations)
- Generalization Bounds for Estimating Causal Effects of Continuous Treatments (5 citations)
- Deep Learning Methods for Proximal Inference via Maximum Moment Restriction (4 citations)
- Deep Multi-Modal Structural Equations For Causal Effect Estimation With Unstructured Proxies (4 citations)
- What's the Harm? Sharp Bounds on the Fraction Negatively Affected by Treatment (4 citations)
- Anticipating Performativity by Predicting from Predictions (3 citations)
- Disentangling Causal Effects from Sets of Interventions in the Presence of Unobserved Confounders (3 citations)
- Sample Constrained Treatment Effect Estimation (3 citations)
- Debiased Causal Tree: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding (2 citations)
and 5 more in the interactive map.
