ICML 2023 accepted papers on Probabilistic and scientific modeling

Variational and generative sampling · 70 papers

• Sampling-Based Accuracy Testing of Posterior Estimators for General Inference (18 citations)
• A Likelihood Approach to Nonparametric Estimation of a Singular Distribution Using Deep Generative Models (13 citations)
• Rigid Body Flows for Sampling Molecular Crystal Structures (10 citations)
• A theory of continuous generative flow networks (5 citations)
• Explaining the effects of non-convergent MCMC in the training of Energy-Based Models (5 citations)
• Minimizing Trajectory Curvature of ODE-based Generative Models (5 citations)
• Vector Quantized Wasserstein Auto-Encoder (5 citations)
• Compositional Score Modeling for Simulation-Based Inference (4 citations)
• Transformed Distribution Matching for Missing Value Imputation (4 citations)
• Variational Autoencoding Neural Operators (4 citations)
• Cooperation in the Latent Space: The Benefits of Adding Mixture Components in Variational Autoencoders (3 citations)
• Multisample Flow Matching: Straightening Flows with Minibatch Couplings (3 citations)

and 58 more in the interactive map.

Uncertainty and Gaussian processes · 38 papers

• Synthetic Data, Real Errors: How (Not) to Publish and Use Synthetic Data (16 citations)
• Estimating the Contamination Factor's Distribution in Unsupervised Anomaly Detection (9 citations)
• Learn to Accumulate Evidence from All Training Samples: Theory and Practice (5 citations)
• Uncertainty Estimation for Molecules: Desiderata and Methods (5 citations)
• End-to-End Learning for Stochastic Optimization: A Bayesian Perspective (4 citations)
• Robust and Scalable Bayesian Online Changepoint Detection (4 citations)
• Gaussian processes at the Helm(holtz): A more fluid model for ocean currents (3 citations)
• GFlowOut: Dropout with Generative Flow Networks (3 citations)
• Masked Bayesian Neural Networks : Theoretical Guarantee and its Posterior Inference (3 citations)
• On Second-Order Scoring Rules for Epistemic Uncertainty Quantification (3 citations)
• Conformal Inference is (almost) Free for Neural Networks Trained with Early Stopping (2 citations)
• Efficient Transformed Gaussian Processes for Non-Stationary Dependent Multi-class Classification (2 citations)

and 26 more in the interactive map.

Molecules and proteins · 36 papers

• ClimaX: A foundation model for weather and climate (167 citations)
• Generating Novel, Designable, and Diverse Protein Structures by Equivariantly Diffusing Oriented Residue Clouds (31 citations)
• DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design (26 citations)
• Chemically Transferable Generative Backmapping of Coarse-Grained Proteins (13 citations)
• End-to-End Full-Atom Antibody Design (12 citations)
• Learning Subpocket Prototypes for Generalizable Structure-based Drug Design (9 citations)
• Spatial Implicit Neural Representations for Global-Scale Species Mapping (9 citations)
• A Group Symmetric Stochastic Differential Equation Model for Molecule Multi-modal Pretraining (6 citations)
• Deep Generative Symbolic Regression with Monte-Carlo-Tree-Search (5 citations)
• Exploring Chemical Space with Score-based Out-of-distribution Generation (5 citations)
• Latent Traversals in Generative Models as Potential Flows (5 citations)
• Differentiable Simulations for Enhanced Sampling of Rare Events (4 citations)

and 24 more in the interactive map.

Differential equations and PDEs · 25 papers

• Learning Physical Models that Can Respect Conservation Laws (21 citations)
• Solving High-Dimensional PDEs with Latent Spectral Models (13 citations)
• Neural Inverse Operators for Solving PDE Inverse Problems (9 citations)
• MultiAdam: Parameter-wise Scale-invariant Optimizer for Multiscale Training of Physics-informed Neural Networks (7 citations)
• Implicit Neural Spatial Representations for Time-dependent PDEs (6 citations)
• Group Equivariant Fourier Neural Operators for Partial Differential Equations (5 citations)
• Random Grid Neural Processes for Parametric Partial Differential Equations (4 citations)
• NeuralStagger: Accelerating Physics-constrained Neural PDE Solver with Spatial-temporal Decomposition (3 citations)
• Disentangled Multi-Fidelity Deep Bayesian Active Learning (2 citations)
• General Covariance Data Augmentation for Neural PDE Solvers (2 citations)
• Learning Neural PDE Solvers with Parameter-Guided Channel Attention (2 citations)
• Locally Regularized Neural Differential Equations: Some Black Boxes were meant to remain closed! (2 citations)

and 13 more in the interactive map.

Dynamical systems · 25 papers

• Geometric Clifford Algebra Networks (10 citations)
• Learning Neural Constitutive Laws from Motion Observations for Generalizable PDE Dynamics (6 citations)
• Flexible Phase Dynamics for Bio-Plausible Contrastive Learning (4 citations)
• ConCerNet: A Contrastive Learning Based Framework for Automated Conservation Law Discovery and Trustworthy Dynamical System Prediction (2 citations)
• Trainability, Expressivity and Interpretability in Gated Neural ODEs (2 citations)
• Action Matching: Learning Stochastic Dynamics from Samples (1 citation)
• Adaptive Whitening in Neural Populations with Gain-modulating Interneurons (1 citation)
• Learning Control-Oriented Dynamical Structure from Data (1 citation)
• System Identification of Neural Systems: If We Got It Right, Would We Know? (1 citation)
• The Acquisition of Physical Knowledge in Generative Neural Networks (1 citation)
• Understanding Self-Predictive Learning for Reinforcement Learning (1 citation)
• Computational Doob h-transforms for Online Filtering of Discretely Observed Diffusions

and 13 more in the interactive map.

Treatment effects · 24 papers

• A Reinforcement Learning Framework for Dynamic Mediation Analysis (2 citations)
• B-Learner: Quasi-Oracle Bounds on Heterogeneous Causal Effects Under Hidden Confounding (2 citations)
• Comparison of meta-learners for estimating multi-valued treatment heterogeneous effects (2 citations)
• In Search of Insights, Not Magic Bullets: Towards Demystification of the Model Selection Dilemma in Heterogeneous Treatment Effect Estimation (2 citations)
• Pricing Experimental Design: Causal Effect, Expected Revenue and Tail Risk (2 citations)
• Proximal Causal Learning of Conditional Average Treatment Effects (2 citations)
• Accounting For Informative Sampling When Learning to Forecast Treatment Outcomes Over Time (1 citation)
• Causal Isotonic Calibration for Heterogeneous Treatment Effects (1 citation)
• Covariate balancing using the integral probability metric for causal inference (1 citation)
• Estimating Causal Effects using a Multi-task Deep Ensemble (1 citation)
• Non-asymptotic Properties of Individualized Treatment Rules from Sequentially Rule-Adaptive Trials (1 citation)
• Normalizing Flows for Interventional Density Estimation (1 citation)

and 12 more in the interactive map.

Causal discovery · 24 papers

• Causal Proxy Models for Concept-based Model Explanations (5 citations)
• High Fidelity Image Counterfactuals with Probabilistic Causal Models (4 citations)
• Interventional Causal Representation Learning (4 citations)
• Linear Causal Disentanglement via Interventions (4 citations)
• On the Relationship Between Explanation and Prediction: A Causal View (4 citations)
• Towards Trustworthy Explanation: On Causal Rationalization (4 citations)
• DiscoBAX - Discovery of optimal intervention sets in genomic experiment design (3 citations)
• On the Identifiability and Estimation of Causal Location-Scale Noise Models (3 citations)
• Causal Discovery with Latent Confounders Based on Higher-Order Cumulants (1 citation)
• Counterfactual Analysis in Dynamic Latent State Models (1 citation)
• Counterfactual Identifiability of Bijective Causal Models (1 citation)
• Approximate Causal Effect Identification under Weak Confounding

and 12 more in the interactive map.

Bayesian optimization · 12 papers

• Meta Learning of Interface Conditions for Multi-Domain Physics-Informed Neural Networks (5 citations)
• Diffusion Models for Black-Box Optimization (4 citations)
• PFNs4BO: In-Context Learning for Bayesian Optimization (3 citations)
• Generative Pretraining for Black-Box Optimization (2 citations)
• Towards Practical Preferential Bayesian Optimization with Skew Gaussian Processes (2 citations)
• Are Random Decompositions all we need in High Dimensional Bayesian Optimisation? (1 citation)
• CO-BED: Information-Theoretic Contextual Optimization via Bayesian Experimental Design (1 citation)
• Constrained Causal Bayesian Optimization (1 citation)
• Delayed Feedback in Kernel Bandits
• Differentiable Multi-Target Causal Bayesian Experimental Design
• Learning Regions of Interest for Bayesian Optimization with Adaptive Level-Set Estimation
• Hypervolume Knowledge Gradient: A Lookahead Approach for Multi-Objective Bayesian Optimization with Partial Information