Navigating Data Corruption in Machine Learning: Balancing Quality, Quantity, and Imputation Strategies
Qi Liu*, Wanjing Ma · 2024
Read Navigating Data Corruption in Machine Learning: Balancing Quality, Quantity, and Imputation Strategies by Qi Liu*, Wanjing Ma (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-18296.txt.
Description
Qi Liu and Wanjing Ma study data corruption in machine learning in two thousand twenty-four. Real data sets carry missing values and noise that hurt models. Their tests span language tasks with supervision and traffic signal control by reinforcement. They weigh fix by filling, and fix by more data.
Missing data hurt less than noisy data in their runs. Noise shook training most in the sequential signal task. Fills help when they stay accurate and hurt when they add noise. More data eases harm but cannot erase it. A rule of thumb puts the key share of data at a small slice.
Catalog record
| CREATOR | Qi Liu*, Wanjing Ma |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-18296