ICML 2022 accepted papers on Probabilistic and scientific modeling
159 of the 1,233 papers accepted at ICML 2022 (12.9%) are on Probabilistic and scientific modeling, compared with 12.4% at ICML 2021. They fall into 8 topics.
See ICML 2022 in the interactive map
Submitting to ICML? 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 · 37 papers
- SQ-VAE: Variational Bayes on Discrete Representation with Self-annealed Stochastic Quantization (13 citations)
- Path-Gradient Estimators for Continuous Normalizing Flows (6 citations)
- Continual Repeated Annealed Flow Transport Monte Carlo (5 citations)
- Generative Flow Networks for Discrete Probabilistic Modeling (5 citations)
- Matching Normalizing Flows and Probability Paths on Manifolds (5 citations)
- Neural Inverse Kinematic (4 citations)
- SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation (4 citations)
- C*-algebra Net: A New Approach Generalizing Neural Network Parameters to C*-algebra (3 citations)
- Variational Sparse Coding with Learned Thresholding (3 citations)
- Fat–Tailed Variational Inference with Anisotropic Tail Adaptive Flows (2 citations)
- Marginal Tail-Adaptive Normalizing Flows (2 citations)
- Markov Chain Monte Carlo for Continuous-Time Switching Dynamical Systems (2 citations)
and 25 more in the interactive map.
Uncertainty and Gaussian processes · 33 papers
- Bayesian Model Selection, the Marginal Likelihood, and Generalization (17 citations)
- Image-to-Image Regression with Distribution-Free Uncertainty Quantification and Applications in Imaging (13 citations)
- Transformer Neural Processes: Uncertainty-Aware Meta Learning Via Sequence Modeling (12 citations)
- On the Practicality of Deterministic Epistemic Uncertainty (6 citations)
- Adapting the Linearised Laplace Model Evidence for Modern Deep Learning (5 citations)
- Additive Gaussian Processes Revisited (5 citations)
- Continual Learning via Sequential Function-Space Variational Inference (5 citations)
- Calibrated and Sharp Uncertainties in Deep Learning via Density Estimation (4 citations)
- Wide Bayesian neural networks have a simple weight posterior: theory and accelerated sampling (4 citations)
- Deep Probability Estimation (3 citations)
- Bayesian Nonparametric Learning for Point Processes with Spatial Homogeneity: A Spatial Analysis of NBA Shot Locations (2 citations)
- How Tempering Fixes Data Augmentation in Bayesian Neural Networks (2 citations)
and 21 more in the interactive map.
Molecules and proteins · 23 papers
- Learning inverse folding from millions of predicted structures (463 citations)
- Tranception: Protein Fitness Prediction with Autoregressive Transformers and Inference-time Retrieval (125 citations)
- EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction (100 citations)
- Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pockets (78 citations)
- De novo mass spectrometry peptide sequencing with a transformer model (61 citations)
- Generating 3D Molecules for Target Protein Binding (46 citations)
- 3DLinker: An E(3) Equivariant Variational Autoencoder for Molecular Linker Design (24 citations)
- HyperImpute: Generalized Iterative Imputation with Automatic Model Selection (24 citations)
- Generative Coarse-Graining of Molecular Conformations (23 citations)
- Biological Sequence Design with GFlowNets (19 citations)
- LIMO: Latent Inceptionism for Targeted Molecule Generation (17 citations)
- Proximal Exploration for Model-guided Protein Sequence Design (17 citations)
and 11 more in the interactive map.
Causal discovery · 19 papers
- Partial Counterfactual Identification from Observational and Experimental Data (10 citations)
- Causal Conceptions of Fairness and their Consequences (8 citations)
- Matching Learned Causal Effects of Neural Networks with Domain Priors (7 citations)
- CITRIS: Causal Identifiability from Temporal Intervened Sequences (6 citations)
- Exploiting Independent Instruments: Identification and Distribution Generalization (5 citations)
- Strategic Instrumental Variable Regression: Recovering Causal Relationships From Strategic Responses (5 citations)
- Inducing Causal Structure for Interpretable Neural Networks (4 citations)
- Causal Inference Through the Structural Causal Marginal Problem (3 citations)
- Score Matching Enables Causal Discovery of Nonlinear Additive Noise Models (2 citations)
- Counterfactual Transportability: A Formal Approach
- Invariant Ancestry Search
- Investigating Generalization by Controlling Normalized Margin
and 7 more in the interactive map.
Dynamical systems · 15 papers
- Constraint-based graph network simulator (10 citations)
- Estimating and Penalizing Induced Preference Shifts in Recommender Systems (10 citations)
- Approximately Equivariant Networks for Imperfectly Symmetric Dynamics (9 citations)
- Variational Mixtures of ODEs for Inferring Cellular Gene Expression Dynamics (7 citations)
- Short-Term Plasticity Neurons Learning to Learn and Forget (6 citations)
- Tractable Dendritic RNNs for Reconstructing Nonlinear Dynamical Systems (6 citations)
- Burst-Dependent Plasticity and Dendritic Amplification Support Target-Based Learning and Hierarchical Imitation Learning (3 citations)
- NeuroFluid: Fluid Dynamics Grounding with Particle-Driven Neural Radiance Fields (3 citations)
- Neural Network Poisson Models for Behavioural and Neural Spike Train Data (2 citations)
- Extracting Latent State Representations with Linear Dynamics from Rich Observations (1 citation)
- Generalizing to New Physical Systems via Context-Informed Dynamics Model (1 citation)
- ModLaNets: Learning Generalisable Dynamics via Modularity and Physical Inductive Bias (1 citation)
and 3 more in the interactive map.
Differential equations and PDEs · 13 papers
- On Numerical Integration in Neural Ordinary Differential Equations (12 citations)
- Lie Point Symmetry Data Augmentation for Neural PDE Solvers (9 citations)
- Learning to Estimate and Refine Fluid Motion with Physical Dynamics (6 citations)
- Composing Partial Differential Equations with Physics-Aware Neural Networks (5 citations)
- Neural Laplace: Learning diverse classes of differential equations in the Laplace domain (5 citations)
- Probabilistic ODE Solutions in Millions of Dimensions (5 citations)
- Continuous-Time Modeling of Counterfactual Outcomes Using Neural Controlled Differential Equations (4 citations)
- Fast Aquatic Swimmer Optimization with Differentiable Projective Dynamics and Neural Network Hydrodynamic Models (4 citations)
- AutoIP: A United Framework to Integrate Physics into Gaussian Processes (3 citations)
- Fenrir: Physics-Enhanced Regression for Initial Value Problems (2 citations)
- Robust SDE-Based Variational Formulations for Solving Linear PDEs via Deep Learning
- Learning Efficient and Robust Ordinary Differential Equations via Invertible Neural Networks
and 1 more in the interactive map.
Bayesian optimization · 11 papers
- Accelerating Bayesian Optimization for Biological Sequence Design with Denoising Autoencoders (22 citations)
- Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimization (7 citations)
- Robust Multi-Objective Bayesian Optimization Under Input Noise (6 citations)
- NOMU: Neural Optimization-based Model Uncertainty (5 citations)
- A General Recipe for Likelihood-free Bayesian Optimization (2 citations)
- Bayesian Optimization for Distributionally Robust Chance-constrained Problem (1 citation)
- Bayesian Optimization under Stochastic Delayed Feedback (1 citation)
- Meta-Learning Hypothesis Spaces for Sequential Decision-making
- Scalable First-Order Bayesian Optimization via Structured Automatic Differentiation
- Distributionally-Aware Kernelized Bandit Problems for Risk Aversion
- Spectral Representation of Robustness Measures for Optimization Under Input Uncertainty
Treatment effects · 8 papers
- Causal Transformer for Estimating Counterfactual Outcomes (18 citations)
- Validating Causal Inference Methods (11 citations)
- A Tree-based Model Averaging Approach for Personalized Treatment Effect Estimation from Heterogeneous Data Sources (9 citations)
- Measuring the Effect of Training Data on Deep Learning Predictions via Randomized Experiments (3 citations)
- End-to-End Balancing for Causal Continuous Treatment-Effect Estimation (2 citations)
- Coordinated Double Machine Learning (1 citation)
- Counterfactual Prediction for Outcome-Oriented Treatments
- Instrumental Variable Regression with Confounder Balancing
