Complex Coordinate-Based Meta-Analysis with Probabilistic Programming

Authors: Valentin Iovene, Gaston E Zanitti, Demian Wassermann223-231

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate results for two-term CQs, both on simulated meta-analysis databases and on the widely used Neurosynth database.
Researcher Affiliation Academia Valentin Iovene, Gaston E Zanitti, Demian Wassermann Inria, CEA, Universit e Paris-Saclay, Palaiseau, France
Pseudocode Yes Figure 1: CP-Logic program encoding a probabilistic CBMA database.
Open Source Code Yes The Neuro Lang program implementing this model is available at https://github.com/Neuro Lang/Neuro Lang/tree/master/examples/plot neurosynth relaxed tfidf.py.
Open Datasets Yes We demonstrate results for two-term CQs, both on simulated meta-analysis databases and on the widely used Neurosynth database.
Dataset Splits No The paper mentions evaluating on 'simulated meta-analysis databases' and the 'Neurosynth database' with varying 'sample sizes' and 'sub-samples', but does not provide specific train/validation/test splits or mention standard predefined splits.
Hardware Specification No The paper states, 'Solving a two-term CQ takes more than 30 minutes on a recent laptop,' but does not provide specific hardware details such as CPU/GPU models or memory.
Software Dependencies No The paper states, 'We implemented the program of fig. 1 in Prob Log2 (Dries et al. 2015),' but does not specify a version number for Prob Log2 or any other key software dependencies.
Experiment Setup Yes The threshold τ = 0.1 is used in both models. The value α = 300 was empirically chosen. Varying α near this value does not change the results noticeably.