NeurIPS 2018 accepted papers on Probabilistic and scientific modeling
164 of the 1,008 papers accepted at NeurIPS 2018 (16.3%) are on Probabilistic and scientific modeling, compared with 18.5% at NeurIPS 2017. They fall into 8 topics.
See NeurIPS 2018 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 · 60 papers
- Glow: Generative Flow with Invertible 1x1 Convolutions (177 citations)
- Assessing Generative Models via Precision and Recall (168 citations)
- Isolating Sources of Disentanglement in Variational Autoencoders (137 citations)
- Generative Probabilistic Novelty Detection with Adversarial Autoencoders (136 citations)
- Implicit Reparameterization Gradients (129 citations)
- Information Constraints on Auto-Encoding Variational Bayes (93 citations)
- A Likelihood-Free Inference Framework for Population Genetic Data using Exchangeable Neural Networks (86 citations)
- Latent Alignment and Variational Attention (79 citations)
- Learning Latent Subspaces in Variational Autoencoders (69 citations)
- Amortized Inference Regularization (67 citations)
- Bias and Generalization in Deep Generative Models: An Empirical Study (67 citations)
- Learning Disentangled Joint Continuous and Discrete Representations (59 citations)
and 48 more in the interactive map.
Uncertainty and Gaussian processes · 34 papers
- Evidential Deep Learning to Quantify Classification Uncertainty (543 citations)
- GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration (528 citations)
- Predictive Uncertainty Estimation via Prior Networks (351 citations)
- Bayesian Model-Agnostic Meta-Learning (156 citations)
- Inference in Deep Gaussian Processes using Stochastic Gradient Hamiltonian Monte Carlo (63 citations)
- Bayesian Control of Large MDPs with Unknown Dynamics in Data-Poor Environments (62 citations)
- Heterogeneous Multi-output Gaussian Process Prediction (48 citations)
- Bayesian Inference of Temporal Task Specifications from Demonstrations (44 citations)
- Scaling Gaussian Process Regression with Derivatives (42 citations)
- Dirichlet-based Gaussian Processes for Large-scale Calibrated Classification (33 citations)
- Explaining Deep Learning Models -- A Bayesian Non-parametric Approach (30 citations)
- Nonparametric learning from Bayesian models with randomized objective functions (29 citations)
and 22 more in the interactive map.
Dynamical systems · 23 papers
- End-to-End Differentiable Physics for Learning and Control (256 citations)
- Flexible neural representation for physics prediction (160 citations)
- Deep Dynamical Modeling and Control of Unsteady Fluid Flows (106 citations)
- Learning Plannable Representations with Causal InfoGAN (83 citations)
- The emergence of multiple retinal cell types through efficient coding of natural movies (47 citations)
- Temporal alignment and latent Gaussian process factor inference in population spike trains (42 citations)
- Deep Generative Markov State Models (41 citations)
- Manifold-tiling Localized Receptive Fields are Optimal in Similarity-preserving Neural Networks (39 citations)
- Efficient inference for time-varying behavior during learning (25 citations)
- Generalisation of structural knowledge in the hippocampal-entorhinal system (23 citations)
- Model-based targeted dimensionality reduction for neuronal population data (22 citations)
- Scaling the Poisson GLM to massive neural datasets through polynomial approximations (22 citations)
and 11 more in the interactive map.
Molecules and proteins · 20 papers
- End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems (218 citations)
- Generative modeling for protein structures (116 citations)
- Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objects (89 citations)
- Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders (79 citations)
- Learning Libraries of Subroutines for Neurally–Guided Bayesian Program Induction (41 citations)
- Modeling Dynamic Missingness of Implicit Feedback for Recommendation (39 citations)
- Scalable End-to-End Autonomous Vehicle Testing via Rare-event Simulation (39 citations)
- Neural Interaction Transparency (NIT): Disentangling Learned Interactions for Improved Interpretability (38 citations)
- Bayesian multi-domain learning for cancer subtype discovery from next-generation sequencing count data (30 citations)
- Inference Aided Reinforcement Learning for Incentive Mechanism Design in Crowdsourcing (21 citations)
- Geometry Based Data Generation (20 citations)
- The Sparse Manifold Transform (17 citations)
and 8 more in the interactive map.
Causal discovery · 14 papers
- Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions (84 citations)
- Equality of Opportunity in Classification: A Causal Approach (32 citations)
- Removing Hidden Confounding by Experimental Grounding (31 citations)
- Causal Inference via Kernel Deviance Measures (30 citations)
- Structural Causal Bandits: Where to Intervene? (29 citations)
- Causal Discovery from Discrete Data using Hidden Compact Representation (25 citations)
- Learning and Testing Causal Models with Interventions (22 citations)
- Multi-domain Causal Structure Learning in Linear Systems (22 citations)
- Identification and Estimation of Causal Effects from Dependent Data (21 citations)
- Causal Inference and Mechanism Clustering of A Mixture of Additive Noise Models (13 citations)
- Bayesian Structure Learning by Recursive Bootstrap (11 citations)
- Fast Estimation of Causal Interactions using Wold Processes (4 citations)
and 2 more in the interactive map.
Bayesian optimization · 7 papers
- Scalable Hyperparameter Transfer Learning (77 citations)
- Maximizing acquisition functions for Bayesian optimization (65 citations)
- Probabilistic Matrix Factorization for Automated Machine Learning (64 citations)
- Automating Bayesian optimization with Bayesian optimization (19 citations)
- Optimization over Continuous and Multi-dimensional Decisions with Observational Data (14 citations)
- Hybrid-MST: A Hybrid Active Sampling Strategy for Pairwise Preference Aggregation (5 citations)
- Scalar Posterior Sampling with Applications (4 citations)
Differential equations and PDEs · 4 papers
- Neural Ordinary Differential Equations (558 citations)
- Neural Networks Trained to Solve Differential Equations Learn General Representations (46 citations)
- Algorithmic Linearly Constrained Gaussian Processes (18 citations)
- Variational PDEs for Acceleration on Manifolds and Application to Diffeomorphisms (7 citations)
