Learning Gaussian Mixture Representations for Tensor Time Series Forecasting

Authors: Jiewen Deng, Jinliang Deng, Renhe Jiang, Xuan Song

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiment results on two real-world TTS datasets verify the superiority of our approach compared with the state-of-the-art baselines. Code and data are published on https://github.com/beginner-sketch/GMRL. We conduct thorough experiments on two real-world datasets.
Researcher Affiliation Academia 1Southern University of Science and Technology 2University of Technology Sydney 3The University of Tokyo
Pseudocode No The paper describes the methodology textually and with mathematical equations, but does not include a distinct pseudocode block or algorithm figure.
Open Source Code Yes Code and data are published on https://github.com/beginner-sketch/GMRL.
Open Datasets Yes We conduct experiments on two real-world TTS datasets, namely NYC Traffic Demand and BJ Air Quality as listed in Table 2, the details of which are as follows: NYC Traffic Demand dataset12 is collected from NYC Bike Sharing System... BJ Air Quality dataset3 is collected from the Beijing Municipal Environmental Monitoring Center... 1https://ride.citibikenyc.com/system-data 2https://www1.nyc.gov/site/tlc/about/tlc-trip-record-data.page 3https://archive.ics.uci.edu/ml/datasets/Beijing+Multi-Site+ Air-Quality+Data
Dataset Splits Yes Table 2: # Train/Val/Test 3912/240/240
Hardware Specification No The paper mentions implementing the network with the Pytorch toolkit, but does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments.
Software Dependencies No The paper states: "We implement the network with the Pytorch toolkit." However, it does not specify any version numbers for PyTorch or other software dependencies.
Experiment Setup Yes For the model, the number of GMRE-TE layers and cluster components K are set to 4 and 17. The kernel size of each dilated causal convolution component is 2, and the related expansion rate is {2, 4, 8, 16} in each GMRE-TE layer. This enables our model to handle the 16 input steps. The dimension of hidden channels dz is 24. The parameters for memory bank m and dm are set to 8 and 48. The batch size is 8, and the learning rate of the Adam optimizer is 0.0001. In addition, the inputs are normalized by Z-Score.