Summary of Probabilistic Forecasting Of Irregular Time Series Via Conditional Flows, by Vijaya Krishna Yalavarthi et al.
Probabilistic Forecasting of Irregular Time Series via Conditional Flows
by Vijaya Krishna Yalavarthi, Randolf Scholz, Stefan Born, Lars Schmidt-Thieme
First submitted to arxiv on: 9 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a new model, ProFITi, for probabilistic forecasting of irregularly sampled multivariate time series with missing values. The model uses conditional normalizing flows to learn joint distributions over future values conditioned on past observations and queried channels and times. The approach is novel in that it doesn’t assume a fixed shape of the underlying distribution. The authors introduce two new components: an invertible triangular attention layer and an invertable non-linear activation function. They conduct experiments on four datasets and show that ProFITi outperforms previous models by a factor of 4. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about predicting things that happen at irregular times with missing data. This is important in fields like healthcare, astronomy, and climate study. Right now, the best methods just predict what will happen next without considering all the other variables. The new model, ProFITi, tries to fix this by looking at everything together and using special math tools to make the predictions more accurate. |
Keywords
* Artificial intelligence * Attention * Time series
