AAAI 2021 accepted papers on Probabilistic and scientific modeling
119 of the 1,654 papers accepted at AAAI 2021 (7.2%) are on Probabilistic and scientific modeling, compared with 5.7% at AAAI 2020. They fall into 8 topics.
See AAAI 2021 in the interactive map
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
Molecules and proteins · 46 papers
- Explaining A Black-box By Using A Deep Variational Information Bottleneck Approach (66 citations)
- GTA: Graph Truncated Attention for Retrosynthesis (62 citations)
- Nested Named Entity Recognition with Partially-Observed TreeCRFs (53 citations)
- Clinical Temporal Relation Extraction with Probabilistic Soft Logic Regularization and Global Inference (47 citations)
- DeepWriteSYN: On-Line Handwriting Synthesis via Deep Short-Term Representations (41 citations)
- Interpreting Multivariate Shapley Interactions in DNNs (36 citations)
- Measuring Dependence with Matrix-based Entropy Functional (31 citations)
- Evidence Inference Networks for Interpretable Claim Verification (29 citations)
- Uncovering Latent Biases in Text: Method and Application to Peer Review (27 citations)
- Ethically Compliant Sequential Decision Making (26 citations)
- Modeling Heterogeneous Relations across Multiple Modes for Potential Crowd Flow Prediction (23 citations)
- A Hybrid Probabilistic Approach for Table Understanding (22 citations)
and 34 more in the interactive map.
Uncertainty and Gaussian processes · 19 papers
- Practical and Rigorous Uncertainty Bounds for Gaussian Process Regression (48 citations)
- MARTA: Leveraging Human Rationales for Explainable Text Classification (42 citations)
- Deep Probabilistic Imaging: Uncertainty Quantification and Multi-modal Solution Characterization for Computational Imaging (39 citations)
- Capturing Uncertainty in Unsupervised GPS Trajectory Segmentation Using Bayesian Deep Learning (25 citations)
- Conceptualized and Contextualized Gaussian Embedding (22 citations)
- DIBS: Diversity Inducing Information Bottleneck in Model Ensembles (21 citations)
- Bayes-TrEx: a Bayesian Sampling Approach to Model Transparency by Example (13 citations)
- Turbocharging Treewidth-Bounded Bayesian Network Structure Learning (11 citations)
- Learning Prediction Intervals for Model Performance (10 citations)
- Uncertainty Quantification in CNN Through the Bootstrap of Convex Neural Networks (10 citations)
- Tracking Disease Outbreaks from Sparse Data with Bayesian Inference (8 citations)
- Vector Quantized Bayesian Neural Network Inference for Data Streams (8 citations)
and 7 more in the interactive map.
Variational and generative sampling · 17 papers
- LREN: Low-Rank Embedded Network for Sample-Free Hyperspectral Anomaly Detection (71 citations)
- OT-Flow: Fast and Accurate Continuous Normalizing Flows via Optimal Transport (57 citations)
- Open-Set Recognition with Gaussian Mixture Variational Autoencoders (35 citations)
- Bayesian Optimized Monte Carlo Planning (20 citations)
- HINT: Hierarchical Invertible Neural Transport for Density Estimation and Bayesian Inference (20 citations)
- Learning Energy-Based Model with Variational Auto-Encoder as Amortized Sampler (20 citations)
- Group Fairness by Probabilistic Modeling with Latent Fair Decisions (14 citations)
- MLE-Guided Parameter Search for Task Loss Minimization in Neural Sequence Modeling (9 citations)
- Improved Mutual Information Estimation (8 citations)
- Deep Probabilistic Canonical Correlation Analysis (6 citations)
- Accelerating Continuous Normalizing Flow with Trajectory Polynomial Regularization (5 citations)
- Variational Inference for Learning Representations of Natural Language Edits (4 citations)
and 5 more in the interactive map.
Dynamical systems · 10 papers
- Differentiable Fluids with Solid Coupling for Learning and Control (18 citations)
- Deep Conservation: A Latent-Dynamics Model for Exact Satisfaction of Physical Conservation Laws (16 citations)
- Neural Latent Space Model for Dynamic Networks and Temporal Knowledge Graphs (15 citations)
- Learning Dynamics Models with Stable Invariant Sets (11 citations)
- Learning Deep Generative Models for Queuing Systems (8 citations)
- Deep Recurrent Belief Propagation Network for POMDPs (7 citations)
- Bayesian Dynamic Mode Decomposition with Variational Matrix Factorization (6 citations)
- Relaxed Clustered Hawkes Process for Student Procrastination Modeling in MOOCs (6 citations)
- Synchronous Dynamical Systems on Directed Acyclic Graphs: Complexity and Algorithms (2 citations)
- Learning Interpretable Models for Coupled Networks Under Domain Constraints (1 citation)
Causal discovery · 9 papers
- Estimating Identifiable Causal Effects through Double Machine Learning (41 citations)
- Counterfactual Fairness with Disentangled Causal Effect Variational Autoencoder (40 citations)
- Improving Causal Discovery By Optimal Bayesian Network Learning (17 citations)
- Testing Independence Between Linear Combinations for Causal Discovery (15 citations)
- Meta Learning for Causal Direction (13 citations)
- Equivalent Causal Models (12 citations)
- Discovering Fully Oriented Causal Networks (11 citations)
- A Generative Adversarial Framework for Bounding Confounded Causal Effects (8 citations)
- PAC Learning of Causal Trees with Latent Variables
Bayesian optimization · 7 papers
- Warm Starting CMA-ES for Hyperparameter Optimization (35 citations)
- MFES-HB: Efficient Hyperband with Multi-Fidelity Quality Measurements (21 citations)
- Mercer Features for Efficient Combinatorial Bayesian Optimization (16 citations)
- Top-k Ranking Bayesian Optimization (13 citations)
- High-Dimensional Bayesian Optimization via Tree-Structured Additive Models (9 citations)
- Evolution Strategies for Approximate Solution of Bayesian Games (7 citations)
- High Dimensional Level Set Estimation with Bayesian Neural Network (6 citations)
Differential equations and PDEs · 6 papers
- Physics-Informed Deep Learning for Traffic State Estimation: A Hybrid Paradigm Informed By Second-Order Traffic Models (103 citations)
- DPM: A Novel Training Method for Physics-Informed Neural Networks in Extrapolation (88 citations)
- ECG ODE-GAN: Learning Ordinary Differential Equations of ECG Dynamics via Generative Adversarial Learning (40 citations)
- Low-Rank Registration Based Manifolds for Convection-Dominated PDEs (18 citations)
- The LOB Recreation Model: Predicting the Limit Order Book from TAQ History Using an Ordinary Differential Equation Recurrent Neural Network (12 citations)
- Differential Spectral Normalization (DSN) for PDE Discovery (8 citations)
Treatment effects · 5 papers
- Treatment Effect Estimation with Disentangled Latent Factors (62 citations)
- Bounding Causal Effects on Continuous Outcome (17 citations)
- Instrumental Variable-based Identification for Causal Effects using Covariate Information (3 citations)
- The Causal Learning of Retail Delinquency (3 citations)
- Learning Adjustment Sets from Observational and Limited Experimental Data (1 citation)
