AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

DAG-Informed Structure Learning from Multi-Dimensional Point Processes

Chunming Zhang, Muhong Gao, Shengji Jia

JMLR 2024 · Probabilistic and scientific modeling › Causal discovery · approach: Differentiable and score-based DAG learning

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the DAG constraint to the L1/weighted-L1 regularized estimator (ablation)nonDAG-L1 / nonDAG-wL1 · Simulated Network 1 (10 nodes, 15 edges)SHD, false positives, FDRImproved
add the DAG constraint (ablation)nonDAG-L1 / nonDAG-wL1 · Simulated Network 1 (10 nodes, 15 edges)Mi (false negatives)No clear effect
add the DAG constraint to the unregularized MLE (ablation)nonDAG-unreg · Simulated Network 2 (10 nodes, 40 edges, dense)Ex, SHD, FDRImproved
add the DAG constraint (ablation)L1 and weighted-L1 estimators · Simulated Network 3 (50 nodes)SHD, FDRImproved
add the DAG constraint (ablation)L1 and weighted-L1 estimators · Simulated Network 4 (100 nodes)SHD, FDRImproved
use adaptive weighted L1 penalty instead of plain L1 (ablation)DAG-constrained estimator · Simulated Networks 1 and 3SHDImproved
add regularization to the unregularized estimator (ablation)nonDAG-unreg / DAG-unreg · Simulated Network 1 (sparse)false positives, SHDImproved
use Flex-AL updates instead of classical AL updatesDAG-wL1 and DAG-unreg · Simulated Network 3, T=1200iterations to convergence: more than 10 -> within 5 stepsImproved
use Flex-AL updates instead of classical AL updatesDAG-wL1 and DAG-unreg · Simulated Network 3, T=1200runtimeImproved
use DAG-wL1 instead of nonDAG-wL1neuronal spike train estimator · Neuronal spike train (EE.188, 102 neurons), testpredictive log-likelihoodImproved
use DAG-wL1 instead of nonDAG-wL1IPTV estimator · IPTV viewing records (81 channels), testpredictive log-likelihoodImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy