ICML 2024 accepted papers on Probabilistic and scientific modeling
320 of the 2,610 papers accepted at ICML 2024 (12.3%) are on Probabilistic and scientific modeling, compared with 13.6% at ICML 2023. They fall into 8 topics.
See ICML 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Variational and generative sampling · 70 papers
- A Computational Framework for Solving Wasserstein Lagrangian Flows (4 citations)
- An amortized approach to non-linear mixed-effects modeling based on neural posterior estimation (4 citations)
- On the Universality of Volume-Preserving and Coupling-Based Normalizing Flows (3 citations)
- A General Theory for Softmax Gating Multinomial Logistic Mixture of Experts (2 citations)
- Sample as you Infer: Predictive Coding with Langevin Dynamics (2 citations)
- Sampling in Unit Time with Kernel Fisher-Rao Flow (2 citations)
- An Interpretable Evaluation of Entropy-based Novelty of Generative Models (1 citation)
- Beyond ELBOs: A Large-Scale Evaluation of Variational Methods for Sampling (1 citation)
- Estimating Unknown Population Sizes Using the Hypergeometric Distribution (1 citation)
- Exact Soft Analytical Side-Channel Attacks using Tractable Circuits (1 citation)
- Identification and Estimation for Nonignorable Missing Data: A Data Fusion Approach (1 citation)
- Iterated Denoising Energy Matching for Sampling from Boltzmann Densities (1 citation)
and 58 more in the interactive map.
Molecules and proteins · 69 papers
- AlphaFold Meets Flow Matching for Generating Protein Ensembles (88 citations)
- Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Design (12 citations)
- MolCRAFT: Structure-Based Drug Design in Continuous Parameter Space (11 citations)
- FlowMM: Generating Materials with Riemannian Flow Matching (9 citations)
- Swallowing the Bitter Pill: Simplified Scalable Conformer Generation (8 citations)
- CasCast: Skillful High-resolution Precipitation Nowcasting via Cascaded Modelling (7 citations)
- EquiPocket: an E(3)-Equivariant Geometric Graph Neural Network for Ligand Binding Site Prediction (7 citations)
- Harmonic Self-Conditioned Flow Matching for joint Multi-Ligand Docking and Binding Site Design (7 citations)
- Position: Is machine learning good or bad for the natural sciences? (7 citations)
- Projecting Molecules into Synthesizable Chemical Spaces (7 citations)
- RNAFlow: RNA Structure & Sequence Design via Inverse Folding-Based Flow Matching (7 citations)
- Multimodal Prototyping for cancer survival prediction (6 citations)
and 57 more in the interactive map.
Uncertainty and Gaussian processes · 45 papers
- Using AI Uncertainty Quantification to Improve Human Decision-Making (6 citations)
- All-in-one simulation-based inference (5 citations)
- Simultaneous identification of models and parameters of scientific simulators (3 citations)
- Variational Learning is Effective for Large Deep Networks (3 citations)
- Improving Neural Additive Models with Bayesian Principles (2 citations)
- A Rate-Distortion View of Uncertainty Quantification (1 citation)
- An Infinite-Width Analysis on the Jacobian-Regularised Training of a Neural Network (1 citation)
- Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shifts (1 citation)
- Enabling Uncertainty Estimation in Iterative Neural Networks (1 citation)
- Experts Don't Cheat: Learning What You Don't Know By Predicting Pairs (1 citation)
- Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods? (1 citation)
- Outlier-robust Kalman Filtering through Generalised Bayes (1 citation)
and 33 more in the interactive map.
Differential equations and PDEs · 37 papers
- Challenges in Training PINNs: A Loss Landscape Perspective (40 citations)
- Transolver: A Fast Transformer Solver for PDEs on General Geometries (11 citations)
- Neural Operators with Localized Integral and Differential Kernels (10 citations)
- Parameterized Physics-informed Neural Networks for Parameterized PDEs (6 citations)
- CoLoRA: Continuous low-rank adaptation for reduced implicit neural modeling of parameterized partial differential equations (3 citations)
- PARCv2: Physics-aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics Modeling (3 citations)
- Stability-Informed Initialization of Neural Ordinary Differential Equations (3 citations)
- Towards General Neural Surrogate Solvers with Specialized Neural Accelerators (3 citations)
- Diffusion Tempering Improves Parameter Estimation with Probabilistic Integrators for Ordinary Differential Equations (2 citations)
- PAPM: A Physics-aware Proxy Model for Process Systems (2 citations)
- Unifying Bayesian Flow Networks and Diffusion Models through Stochastic Differential Equations (2 citations)
- Accelerating Legacy Numerical Solvers by Non-intrusive Gradient-based Meta-solving (1 citation)
and 25 more in the interactive map.
Causal discovery · 35 papers
- Stable Differentiable Causal Discovery (4 citations)
- Jacobian Regularizer-based Neural Granger Causality (2 citations)
- Predictive Coding beyond Correlations (2 citations)
- Adaptive Online Experimental Design for Causal Discovery (1 citation)
- Causal Discovery with Fewer Conditional Independence Tests (1 citation)
- Causal Effect Identification in LiNGAM Models with Latent Confounders (1 citation)
- Causal Representation Learning from Multiple Distributions: A General Setting (1 citation)
- Challenges and Considerations in the Evaluation of Bayesian Causal Discovery (1 citation)
- Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inference (1 citation)
- On the Recoverability of Causal Relations from Temporally Aggregated I.I.D. Data (1 citation)
- Optimal Transport for Structure Learning Under Missing Data (1 citation)
- A Fixed-Point Approach for Causal Generative Modeling
and 23 more in the interactive map.
Treatment effects · 29 papers
- Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction (2 citations)
- Hybrid$^2$ Neural ODE Causal Modeling and an Application to Glycemic Response (2 citations)
- Active Adaptive Experimental Design for Treatment Effect Estimation with Covariate Choice (1 citation)
- Allocation Requires Prediction Only if Inequality Is Low
- Automating the Selection of Proxy Variables of Unmeasured Confounders
- Causal Inference from Competing Treatments
- Collaborative Heterogeneous Causal Inference Beyond Meta-analysis
- Continuous Treatment Effects with Surrogate Outcomes
- DISCRET: Synthesizing Faithful Explanations For Treatment Effect Estimation
- Doubly Robust Causal Effect Estimation under Networked Interference via Targeted Learning
- Inferring the Long-Term Causal Effects of Long-Term Treatments from Short-Term Experiments
- Learning Decision Policies with Instrumental Variables through Double Machine Learning
and 17 more in the interactive map.
Dynamical systems · 20 papers
- DySLIM: Dynamics Stable Learning by Invariant Measure for Chaotic Systems (4 citations)
- Equivariant Graph Neural Operator for Modeling 3D Dynamics (3 citations)
- Harnessing the Power of Neural Operators with Automatically Encoded Conservation Laws (2 citations)
- Out-of-Domain Generalization in Dynamical Systems Reconstruction (2 citations)
- Latent Space Symmetry Discovery (1 citation)
- A sampling theory perspective on activations for implicit neural representations
- Amortized Equation Discovery in Hybrid Dynamical Systems
- convSeq: Fast and Scalable Method for Detecting Patterns in Spike Data
- Discovering Symmetry Breaking in Physical Systems with Relaxed Group Convolution
- Exploring the Enigma of Neural Dynamics Through A Scattering-Transform Mixer Landscape for Riemannian Manifold
- HelmFluid: Learning Helmholtz Dynamics for Interpretable Fluid Prediction
- MorphGrower: A Synchronized Layer-by-layer Growing Approach for Plausible Neuronal Morphology Generation
and 8 more in the interactive map.
Bayesian optimization · 15 papers
- Vanilla Bayesian Optimization Performs Great in High Dimensions (8 citations)
- Boundary Exploration for Bayesian Optimization With Unknown Physical Constraints (2 citations)
- In-Context Freeze-Thaw Bayesian Optimization for Hyperparameter Optimization (2 citations)
- Probability Distribution of Hypervolume Improvement in Bi-objective Bayesian Optimization (2 citations)
- Accelerating Look-ahead in Bayesian Optimization: Multilevel Monte Carlo is All you Need (1 citation)
- Bayesian Optimization of Function Networks with Partial Evaluations (1 citation)
- BOtied: Multi-objective Bayesian optimization with tied multivariate ranks (1 citation)
- MALIBO: Meta-learning for Likelihood-free Bayesian Optimization (1 citation)
- Boosting Offline Optimizers with Surrogate Sensitivity
- Joint Composite Latent Space Bayesian Optimization
- Learning Surrogates for Offline Black-Box Optimization via Gradient Matching
- Principled Preferential Bayesian Optimization
and 3 more in the interactive map.