Learning Decomposed Spatial Relations for Multi-Variate Time-Series Modeling
Authors: Yuchen Fang, Kan Ren, Caihua Shan, Yifei Shen, You Li, Weinan Zhang, Yong Yu, Dongsheng Li
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The experimental results show that our proposed model outperforms the state-of-the-art baseline methods on both time-series forecasting and time-series point prediction tasks. |
| Researcher Affiliation | Collaboration | 1Shanghai Jiao Tong University, 2Microsoft Research Asia, 3Central South University |
| Pseudocode | No | The paper describes the proposed method using mathematical formulas and descriptive text but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | The supplementary materials and codes are available online1. 1https://seqml.github.io/srd |
| Open Datasets | Yes | Solar (Lai et al. 2018): The solar power production records. Electricity (Lai et al. 2018): The hourly electricity consumption in k Wh. Pems-bay (Li et al. 2017): The average traffic speed in the Bay Area. Metr-la (Li et al. 2017): The average traffic speed measured on the highways of Los Angeles County. Neonatal seizure detection (Stevenson et al. 2019) is a multi-channel electroencephalography (EEG) recording dataset from human neonates for seizure event detection. |
| Dataset Splits | Yes | We split time-series forecasting datasets into training set, validation set and test set following (Wu et al. 2020). And cross-validation has been conducted on the point prediction dataset. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper mentions temporal modules like GRU and TCN and various baseline models, but it does not specify software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions or specific library versions). |
| Experiment Setup | No | The paper states 'determining the best hyper-parameters' and refers to 'α1 and α2 are the hyper-parameters controlling the distinguishability', but does not explicitly list the specific values of these or other common hyperparameters (e.g., learning rate, batch size, number of epochs) used in the experimental setup. |