FaDIn: Fast Discretized Inference for Hawkes Processes with General Parametric Kernels

Authors: Guillaume Staerman, Cédric Allain, Alexandre Gramfort, Thomas Moreau

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental the statistical and computational efficiency of the novel approach is demonstrated through various numerical experiments. Finally, the method s effectiveness is evaluated by modeling the occurrence of stimuli-induced patterns from brain signals recorded with magnetoencephalography (MEG).
Researcher Affiliation Academia 1Universit e Paris-Saclay, Inria, CEA, Palaiseau, 91120, France. Correspondence to: Guillaume Staerman <guillaume.staerman@inria.fr>.
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper mentions using and implementing some features from third-party libraries (tick, pyhawkes, alphacsc) but does not provide a concrete statement or link for the open-source code of their proposed Fa DIn method.
Open Datasets Yes Experiments on MEG data were run on two datasets from the MNE Python package (Gramfort et al., 2013; 2014): the sample dataset and the somatosensory (somato) dataset2. 2Both available at https://mne.tools/stable/ overview/datasets_index.html
Dataset Splits No The paper does not provide specific details on how the datasets were split into training, validation, or test sets, such as percentages or sample counts.
Hardware Specification No The paper does not provide specific hardware details such as GPU or CPU models, memory specifications, or cloud computing instance types used for running the experiments.
Software Dependencies No The paper mentions several software packages (tick, pyhawkes, alphacsc, MNE Python) and algorithms (RMSprop) used in the experiments but does not specify their version numbers.
Experiment Setup Yes The parameter W is set to 1. This experiment is repeated for varying values of T {103, 104, 105}. ... The parameter W of Fa DIn is set to 1. ... The RMSprop algorithm is used in Fa DIn. ... The discretization size of the non-parametric kernel is settled as in Fa DIn. ... with a grid discretization equal to data re-sampling rate of 150 Hz (i.e., δ = 1/150).