REVIEW 10 cited by
Gradient-Based Neural DAG Learning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
We propose a novel score-based approach to learning a directed acyclic graph (DAG) from observational data. We adapt a recently proposed continuous constrained optimization formulation to allow for nonlinear relationships between variables using neural networks. This extension allows to model complex interactions while avoiding the combinatorial nature of the problem. In addition to comparing our method to existing continuous optimization methods, we provide missing empirical comparisons to nonlinear greedy search methods. On both synthetic and real-world data sets, this new method outperforms current continuous methods on most tasks, while being competitive with existing greedy search methods on important metrics for causal inference.
Forward citations
Cited by 10 Pith papers
-
SVI-DAG: A Structured Variational Inference Approach to Bayesian Causal Discovery
SVI-DAG couples normalizing flows over edge logits with stein variational gradient descent on node orderings to learn multimodal Bayesian posteriors over DAGs.
-
polyDAG: Polynomial Acyclicity Constraints for Efficient Continuous Causal Discovery in Visual Semantic Graphs
polyDAG replaces the matrix-exponential acyclicity constraint with a finite polynomial trace constraint proven to be zero exactly on acyclic graphs, plus a geometric-series implementation, yielding faster runtime and ...
-
Toward Temporal Causal Representation Learning with Tensor Decomposition
CaRTeD jointly learns latent phenotypes and the temporal causal network among them from irregular tensors like EHR data, with a convergence guarantee under strong assumptions.
-
Learning Causal Graphs at Scale: A Foundation Model Approach
ADAG pre-trains a linear transformer to map observed data from many related tasks directly to DAG adjacency matrices, enabling fast zero-shot causal discovery on new order-consistent or heterogeneous datasets.
-
Learning General Causal Structures with Hidden Dynamic Process for Climate Analysis
CaDRe jointly recovers latent dynamic processes and observed causal graphs from time-series data, with identifiability theory and competitive climate forecasting.
-
Fast Causal Discovery by Approximate Kernel-based Generalized Score Functions with Linear Computational Complexity
A low-rank approximation of kernel-based generalized score functions reduces the computation and memory cost of score-based causal discovery to linear in sample size with comparable accuracy.
-
CauScale: Neural Causal Discovery at Scale
CauScale uses a two-stream neural architecture with a sample-reduction unit and tied attention weights to scale amortized causal discovery to 1000-node graphs.
-
Scalable Temporal Anomaly Causality Discovery in Large Systems: Achieving Computational Efficiency with Binary Anomaly Flag Data
Temporal causal graphs can be learned from binary alarm flags with 99% data compression and modest accuracy gains using heuristic modifications to PCMCI.
-
From Images to Insights: Explainable Biodiversity Monitoring with Plain Language Habitat Explanations
An image-to-text pipeline that combines species recognition, occurrence data, and causal inference to produce plain-language habitat preference explanations, demonstrated on one bee and one flower species.
-
Graph-attention-based Casual Discovery with Trust Region-navigated Clipping Policy Optimization
The paper claims that trust-region-triggered clipping policy optimization with a scaled dot-product graph attention encoder improves RL-based causal discovery on synthetic and benchmark datasets.
Discussion (0). Continue with ORCID to comment.