AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Granger Causal Inference in Multivariate Hawkes Processes by Minimum Message Length

Katerina Hlaváčková-Schindler, Anna Melnykova, Irene Tubikanec

JMLR 2024 · Probabilistic and scientific modeling › Causal discovery · approach: Causal discovery from time series · 1 citations (OpenAlex)

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
use MMLH-u instead of BICMMLH · Cascade MHP, p=7, T=200F1 score: 0.943 -> 0.948Improved
use MMLH-u instead of AICMMLH · Cascade MHP, p=20, T=200F1 score: 0.719 -> 0.933Improved
use MMLH-e instead of MMLH-u priorMMLH · Sparse settingsF1 score: 0.948 -> 0.948No clear effect
increase time horizon T from 200 to 700MMLH-u · Cascade MHP, p=7F1 score: 0.948 -> 0.985Improved
increase time horizon T from 200 to 1400MMLH-u · Mid-dense Bernoulli MHP, p=7F1 score: 0.567 -> 0.872Improved
use MDLH instead of MMLH-uMMLH-u · Mid-dense Bernoulli MHP, p=7, T=200F1 score: 0.567 -> 0.767Hurt
use MDLH instead of MMLH-uMMLH-u · Single-input MHP, p=7, T=200F1 score: 0.956 -> 0.768Hurt
decrease uniform prior b (increase exponential c) (ablation)MMLH · Cascade MHP, p=7, T=200F1 scoreHurt
decrease b to 4 and increase c to 0.3 (weaker sparsity penalty) (ablation)MMLH · Mid-dense Bernoulli MHP, p=7, T=200F1 score: 0.399 -> 0.567Improved
increase decay constant beta from 1 to 2MMLH-u · Cascade MHP, p=7, T=200F1 score: 0.948 -> 0.715Hurt
increase T to 324 to compensate beta=2MMLH-u · Cascade MHP, p=7, beta=2F1 score: 0.715 -> 0.798Mixed
use reduced structure set (m=3) with p=20MMLH-u · Cascade MHP, p=20, T=200F1 score: 0.948 -> 0.933No clear effect

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