FITS: Modeling Time Series with $10k$ Parameters
Authors: Zhijian Xu, Ailing Zeng, Qiang Xu
ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We present the results of our experiments on long-term forecasting in Tab. 1 and Tab. 2. The results for short-term forecasting on the M4 dataset are provided in the Appendix. We conduct a comprehensive case study on the performance of FITS using the ETTh2 dataset, which further highlights the impact of the lookback window and cutoff frequency on model performance. |
| Researcher Affiliation | Collaboration | 1Department of Computer Science and Engineering, CUHK 2International Digital Economy Academy (IDEA) |
| Pseudocode | No | The paper includes pipeline diagrams (Figure 2 and Figure 4) but no explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code is available at: https://github.com/VEWOXIC/FITS. |
| Open Datasets | Yes | All datasets used in our experiments are widely-used and publicly available real-world datasets, including, Traffic, Electricity, Weather, ETT (Zhou et al., 2021). We summarize the characteristics of these datasets in appendix. We use five commonly used benchmark datasets: SMD (Server Machine Dataset (Su et al., 2019)), PSM (Polled Server Metrics (Abdulaal et al., 2021)), SWa T (Secure Water Treatment (Mathur & Tippenhauer, 2016)), MSL (Mars Science Laboratory rover), and SMAP (Soil Moisture Active Passive satellite) (Hundman et al., 2018). |
| Dataset Splits | Yes | To avoid information leakage, We choose the hyper-parameter based on the performance of the validation set. The threshold is selected based on the highest F1 score achieved on the validation set. |
| Hardware Specification | No | The paper mentions 'edge devices' and discusses computational efficiency (MACs), but does not specify the exact hardware (e.g., GPU models, CPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not list specific version numbers for software dependencies such as programming languages, libraries, or frameworks (e.g., Python version, PyTorch/TensorFlow version). |
| Experiment Setup | Yes | We conduct grid search on the look-back window of 90, 180, 360, 720 and cutoff frequency, the only hyper-parameter. Further experiments also show that a longer look-back window can result in better performance in most cases. To avoid information leakage, We choose the hyper-parameter based on the performance of the validation set. We use a window size of 200 and downsample the time series segment by a factor of 4 as the input to train FITS to reconstruct the original segment. |