ICML 2020 accepted papers on Probabilistic and scientific modeling
171 of the 1,087 papers accepted at ICML 2020 (15.7%) are on Probabilistic and scientific modeling, compared with 16.5% at ICML 2019. They fall into 8 topics.
See ICML 2020 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 · 74 papers
- Normalizing Flows on Tori and Spheres (71 citations)
- Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densities (66 citations)
- On Variational Learning of Controllable Representations for Text without Supervision (38 citations)
- Semi-Supervised Learning with Normalizing Flows (36 citations)
- Optimal transport mapping via input convex neural networks (35 citations)
- Learning the Stein Discrepancy for Training and Evaluating Energy-Based Models without Sampling (32 citations)
- Likelihood-free MCMC with Amortized Approximate Ratio Estimators (27 citations)
- Einsum Networks: Fast and Scalable Learning of Tractable Probabilistic Circuits (26 citations)
- The Usual Suspects? Reassessing Blame for VAE Posterior Collapse (26 citations)
- Approximation Capabilities of Neural ODEs and Invertible Residual Networks (25 citations)
- Tails of Lipschitz Triangular Flows (22 citations)
- On Contrastive Learning for Likelihood-free Inference (21 citations)
and 62 more in the interactive map.
Uncertainty and Gaussian processes · 36 papers
- Uncertainty Estimation Using a Single Deep Deterministic Neural Network (275 citations)
- NGBoost: Natural Gradient Boosting for Probabilistic Prediction (148 citations)
- SDE-Net: Equipping Deep Neural Networks with Uncertainty Estimates (60 citations)
- Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networks (34 citations)
- How Good is the Bayes Posterior in Deep Neural Networks Really? (32 citations)
- Efficient and Scalable Bayesian Neural Nets with Rank-1 Factors (29 citations)
- The k-tied Normal Distribution: A Compact Parameterization of Gaussian Mean Field Posteriors in Bayesian Neural Networks (21 citations)
- Bayesian Learning from Sequential Data using Gaussian Processes with Signature Covariances (17 citations)
- Randomly Projected Additive Gaussian Processes for Regression (16 citations)
- Scalable Exact Inference in Multi-Output Gaussian Processes (16 citations)
- Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift (16 citations)
- Sparse Gaussian Processes with Spherical Harmonic Features (14 citations)
and 24 more in the interactive map.
Molecules and proteins · 19 papers
- Multi-Objective Molecule Generation using Interpretable Substructures (94 citations)
- A Graph to Graphs Framework for Retrosynthesis Prediction (84 citations)
- WaveFlow: A Compact Flow-based Model for Raw Audio (53 citations)
- An EM Approach to Non-autoregressive Conditional Sequence Generation (28 citations)
- Efficient nonparametric statistical inference on population feature importance using Shapley values (18 citations)
- Towards Non-Parametric Drift Detection via Dynamic Adapting Window Independence Drift Detection (DAWIDD) (13 citations)
- Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis (11 citations)
- CAUSE: Learning Granger Causality from Event Sequences using Attribution Methods (7 citations)
- The Differentiable Cross-Entropy Method (6 citations)
- A Flexible Framework for Nonparametric Graphical Modeling that Accommodates Machine Learning (4 citations)
- Inverse Active Sensing: Modeling and Understanding Timely Decision-Making (4 citations)
- Aggregation of Multiple Knockoffs (3 citations)
and 7 more in the interactive map.
Causal discovery · 13 papers
- Strategic Classification is Causal Modeling in Disguise (33 citations)
- DeepMatch: Balancing Deep Covariate Representations for Causal Inference Using Adversarial Training (21 citations)
- Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders (14 citations)
- Full Law Identification in Graphical Models of Missing Data: Completeness Results (7 citations)
- Causal Effect Identifiability under Partial-Observability (6 citations)
- LazyIter: A Fast Algorithm for Counting Markov Equivalent DAGs and Designing Experiments (6 citations)
- Distinguishing Cause from Effect Using Quantiles: Bivariate Quantile Causal Discovery (5 citations)
- Causal Structure Discovery from Distributions Arising from Mixtures of DAGs (4 citations)
- Characterizing Distribution Equivalence and Structure Learning for Cyclic and Acyclic Directed Graphs (3 citations)
- Cost-effectively Identifying Causal Effects When Only Response Variable is Observable (3 citations)
- Efficient Intervention Design for Causal Discovery with Latents (3 citations)
- Learning and Sampling of Atomic Interventions from Observations (3 citations)
and 1 more in the interactive map.
Dynamical systems · 10 papers
- TrajectoryNet: A Dynamic Optimal Transport Network for Modeling Cellular Dynamics (44 citations)
- SimGANs: Simulator-Based Generative Adversarial Networks for ECG Synthesis to Improve Deep ECG Classification (30 citations)
- Scalable Differentiable Physics for Learning and Control (27 citations)
- Causal Modeling for Fairness In Dynamical Systems (22 citations)
- Collapsed Amortized Variational Inference for Switching Nonlinear Dynamical Systems (11 citations)
- Interference and Generalization in Temporal Difference Learning (10 citations)
- Boosting for Control of Dynamical Systems (5 citations)
- Predictive Coding for Locally-Linear Control (5 citations)
- Scalable Identification of Partially Observed Systems with Certainty-Equivalent EM (1 citation)
- A general recurrent state space framework for modeling neural dynamics during decision-making
Bayesian optimization · 9 papers
- Multi-fidelity Bayesian Optimization with Max-value Entropy Search and its Parallelization (23 citations)
- Population-Based Black-Box Optimization for Biological Sequence Design (15 citations)
- A Quantile-based Approach for Hyperparameter Transfer Learning (14 citations)
- Multi-objective Bayesian Optimization using Pareto-frontier Entropy (13 citations)
- Bayesian Optimisation over Multiple Continuous and Categorical Inputs (12 citations)
- Knowing The What But Not The Where in Bayesian Optimization (11 citations)
- Projective Preferential Bayesian Optimization (3 citations)
- Modulating Surrogates for Bayesian Optimization (1 citation)
- Optimizing Dynamic Structures with Bayesian Generative Search
Differential equations and PDEs · 6 papers
- Combining Differentiable PDE Solvers and Graph Neural Networks for Fluid Flow Prediction (159 citations)
- How to Train Your Neural ODE: the World of Jacobian and Kinetic Regularization (92 citations)
- Amortized Finite Element Analysis for Fast PDE-Constrained Optimization (9 citations)
- Transformation of ReLU-based recurrent neural networks from discrete-time to continuous-time (7 citations)
- Stochastic Differential Equations with Variational Wishart Diffusions (6 citations)
- Learning Similarity Metrics for Numerical Simulations (2 citations)
Treatment effects · 4 papers
- Causal Inference using Gaussian Processes with Structured Latent Confounders (12 citations)
- Causal Effect Estimation and Optimal Dose Suggestions in Mobile Health (4 citations)
- Estimating the Number and Effect Sizes of Non-null Hypotheses (4 citations)
- Counterfactual Cross-Validation: Stable Model Selection Procedure for Causal Inference Models (3 citations)
