ICML 2019 accepted papers on Probabilistic and scientific modeling
128 of the 774 papers accepted at ICML 2019 (16.5%) are on Probabilistic and scientific modeling, compared with 17.0% at ICML 2018. They fall into 8 topics.
See ICML 2019 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 · 62 papers
- Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design (206 citations)
- Invertible Residual Networks (175 citations)
- On Variational Bounds of Mutual Information (143 citations)
- MIWAE: Deep Generative Modelling and Imputation of Incomplete Data Sets (112 citations)
- Automatic Posterior Transformation for Likelihood-Free Inference (69 citations)
- A Wrapped Normal Distribution on Hyperbolic Space for Gradient-Based Learning (61 citations)
- Mixture Models for Diverse Machine Translation: Tricks of the Trade (59 citations)
- Bit-Swap: Recursive Bits-Back Coding for Lossless Compression with Hierarchical Latent Variables (56 citations)
- Disentangling Disentanglement in Variational Autoencoders (51 citations)
- Conditioning by adaptive sampling for robust design (45 citations)
- Latent Normalizing Flows for Discrete Sequences (43 citations)
- Understanding and Accelerating Particle-Based Variational Inference (41 citations)
and 50 more in the interactive map.
Uncertainty and Gaussian processes · 27 papers
- Rates of Convergence for Sparse Variational Gaussian Process Regression (48 citations)
- Understanding Priors in Bayesian Neural Networks at the Unit Level (34 citations)
- Dropout as a Structured Shrinkage Prior (23 citations)
- Scalable Nonparametric Sampling from Multimodal Posteriors with the Posterior Bootstrap (22 citations)
- Deep Gaussian Processes with Importance-Weighted Variational Inference (17 citations)
- Bayesian leave-one-out cross-validation for large data (15 citations)
- Bayesian Joint Spike-and-Slab Graphical Lasso (13 citations)
- Partially Exchangeable Networks and Architectures for Learning Summary Statistics in Approximate Bayesian Computation (13 citations)
- Good Initializations of Variational Bayes for Deep Models (10 citations)
- Distribution calibration for regression (9 citations)
- Minimal Achievable Sufficient Statistic Learning (7 citations)
- The Kernel Interaction Trick: Fast Bayesian Discovery of Pairwise Interactions in High Dimensions (6 citations)
and 15 more in the interactive map.
Molecules and proteins · 11 papers
- Doubly Robust Joint Learning for Recommendation on Data Missing Not at Random (96 citations)
- CHiVE: Varying Prosody in Speech Synthesis with a Linguistically Driven Dynamic Hierarchical Conditional Variational Network (52 citations)
- Molecular Hypergraph Grammar with Its Application to Molecular Optimization (50 citations)
- EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAE (43 citations)
- Learning to Groove with Inverse Sequence Transformations (40 citations)
- Imputing Missing Events in Continuous-Time Event Streams (9 citations)
- Decomposing feature-level variation with Covariate Gaussian Process Latent Variable Models (8 citations)
- kernelPSI: a Post-Selection Inference Framework for Nonlinear Variable Selection (6 citations)
- Functional Transparency for Structured Data: a Game-Theoretic Approach (4 citations)
- Neurally-Guided Structure Inference (3 citations)
- Stochastic Deep Networks (1 citation)
Causal discovery · 7 papers
- Neural Network Attributions: A Causal Perspective (32 citations)
- Sensitivity Analysis of Linear Structural Causal Models (26 citations)
- Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models (22 citations)
- Robustly Disentangled Causal Mechanisms: Validating Deep Representations for Interventional Robustness (21 citations)
- Causal Identification under Markov Equivalence: Completeness Results (20 citations)
- Learning Hawkes Processes Under Synchronization Noise (9 citations)
- Adjustment Criteria for Generalizing Experimental Findings (4 citations)
Dynamical systems · 6 papers
- A Dynamical Systems Perspective on Nesterov Acceleration (49 citations)
- Learning interpretable continuous-time models of latent stochastic dynamical systems (40 citations)
- Switching Linear Dynamics for Variational Bayes Filtering (32 citations)
- Efficient Amortised Bayesian Inference for Hierarchical and Nonlinear Dynamical Systems (10 citations)
- A Recurrent Neural Cascade-based Model for Continuous-Time Diffusion (8 citations)
- Overcoming Mean-Field Approximations in Recurrent Gaussian Process Models (7 citations)
Bayesian optimization · 6 papers
- A Framework for Bayesian Optimization in Embedded Subspaces (55 citations)
- Teaching a black-box learner (25 citations)
- Gaining Free or Low-Cost Interpretability with Interpretable Partial Substitute (23 citations)
- Asynchronous Batch Bayesian Optimisation with Improved Local Penalisation (17 citations)
- Myopic Posterior Sampling for Adaptive Goal Oriented Design of Experiments (13 citations)
- Bayesian Optimization of Composite Functions (6 citations)
Treatment effects · 6 papers
- Validating Causal Inference Models via Influence Functions (36 citations)
- Orthogonal Random Forest for Causal Inference (31 citations)
- Active Learning for Decision-Making from Imbalanced Observational Data (15 citations)
- Learning Models from Data with Measurement Error: Tackling Underreporting (3 citations)
- Classifying Treatment Responders Under Causal Effect Monotonicity (2 citations)
- Inferring Heterogeneous Causal Effects in Presence of Spatial Confounding (2 citations)
Differential equations and PDEs · 3 papers
- Learning to Optimize Multigrid PDE Solvers (29 citations)
- AReS and MaRS Adversarial and MMD-Minimizing Regression for SDEs (11 citations)
- GOODE: A Gaussian Off-The-Shelf Ordinary Differential Equation Solver (7 citations)
