Summary of Rs-reg: Probabilistic and Robust Certified Regression Through Randomized Smoothing, by Aref Miri Rekavandi et al.
RS-Reg: Probabilistic and Robust Certified Regression Through Randomized Smoothing
by Aref Miri Rekavandi, Olga Ohrimenko, Benjamin I.P. Rubinstein
First submitted to arxiv on: 14 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 Randomized smoothing has shown promising certified robustness against adversaries in classification tasks. By defining robustness in regression tasks flexibly through probabilities, the paper demonstrates how to establish upper bounds on input data point perturbation (using the _2 norm) for a user-specified probability of observing valid outputs. The authors showcase the asymptotic property of a basic averaging function in scenarios where the regression model operates without any constraint and derive a certified upper bound of the input perturbations when dealing with a family of regression models where the outputs are bounded. Simulations verify the validity of the theoretical results, revealing the advantages and limitations of simple smoothing functions, such as averaging, in regression tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Randomized smoothing has been successful in classification tasks, but it hasn’t been applied to regression tasks yet. This paper shows how to make regression models more robust by adding some noise to the data. It does this by defining what “robust” means for regression tasks and then finding a limit on how much you can change the input data before the output is still valid. The paper also looks at how well simple smoothing functions work in these situations. |
Keywords
» Artificial intelligence » Classification » Probability » Regression
