Summary of Robust Deep Hawkes Process Under Label Noise Of Both Event and Occurrence, by Xiaoyu Tan et al.
Robust Deep Hawkes Process under Label Noise of Both Event and Occurrence
by Xiaoyu Tan, Bin Li, Xihe Qiu, Jingjing Huang, Yinghui Xu, Wei Chu
First submitted to arxiv on: 24 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 Integrating deep neural networks with the Hawkes process has led to significant improvements in predictive capabilities across finance, health informatics, and information technology. However, these models often struggle in real-world settings due to substantial label noise. This is particularly concerning in healthcare, where misdiagnoses or delayed updates can lead to increased prediction risks. Our research reveals that deep Hawkes process models exhibit reduced robustness when faced with label noise affecting both event types and timing. To address this challenge, we propose the Robust Deep Hawkes Process (RDHP) framework, which considers both events and their occurrences in approximated intensity functions. We tested RDHP on open-source benchmarks with synthetic noise and conducted a case study on obstructive sleep apnea-hypopnea syndrome (OSAHS) in a real-world setting with inherent label noise. The results demonstrate that RDHP can effectively perform classification and regression tasks, even in the presence of label noise related to events and their timing. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research helps machines make better predictions by combining deep neural networks with a special process called the Hawkes process. This combination has been successful in many fields, but it struggles when there’s a lot of incorrect information (label noise). In healthcare, this can be very important because misdiagnoses or delays can lead to bad outcomes. The researchers found that these models don’t work well when they’re faced with both incorrect event types and timing. To solve this problem, they created a new framework called the Robust Deep Hawkes Process (RDHP) that takes into account both events and their times. They tested RDHP on some real-world data and showed that it can make accurate predictions even with label noise. |
Keywords
» Artificial intelligence » Classification » Regression
