Temporal Graph ODEs for Irregularly-Sampled Time Series

Authors: Alessio Gravina, Daniele Zambon, Davide Bacciu, Cesare Alippi

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically validate the proposed approach on several graph benchmarks, showing that TG-ODE can achieve state-of-the-art performance in irregular graph stream tasks.
Researcher Affiliation Academia 1University of Pisa, Pisa, Italy 2The Swiss AI Lab IDSIA, Universit a della Svizzera italiana, Lugano, Switzerland 3Politecnico di Milano, Milan, Italy
Pseudocode No The paper includes mathematical equations but does not present any pseudocode or clearly labeled algorithm blocks.
Open Source Code Yes We release the code implementing our methodology and reproducing our empirical analysis at https://github.com/gravins/TG-ODE.
Open Datasets Yes We considered six real-world graph benchmarks for traffic forecasting: Metr LA [Li et al., 2018], Montevideo [Rozemberczki et al., 2021], Pe MS03 [Guo et al., 2022], Pe MS04 [Guo et al., 2022], Pe MS07 [Guo et al., 2022], and Pe MS08 [Guo et al., 2022]; we report additional details about the datasets in Table 4.
Dataset Splits Yes The training set consists of 100 randomly selected timestamps over the 1000 steps used to simulate the diffusion process. The validation and test sets are generated from two different simulations similar to the one used for building the training set.
Hardware Specification Yes We carried out the experiments on 7 nodes of a cluster with 96 CPUs per node. (...) Average time per epoch (measured in seconds) and std computed using an Intel Xeon Gold 6240R CPU @ 2.40GHz.
Software Dependencies No The paper does not provide specific version numbers for software dependencies (e.g., Python, PyTorch, TensorFlow, etc.).
Experiment Setup Yes We report in Table 1 the grid of hyper-parameters employed in our experiments by each method.