1 option
Adversarial Robustness for Estimation and Alignment / Patrick Chao.
- Format:
- Book
- Thesis/Dissertation
- Author/Creator:
- Chao, Patrick, author.
- Language:
- English
- Subjects (All):
- Statistics.
- Information technology.
- Statistics and Data Science--Penn dissertations.
- Penn dissertations--Statistics and Data Science.
- Local Subjects:
- Statistics.
- Information technology.
- Statistics and Data Science--Penn dissertations.
- Penn dissertations--Statistics and Data Science.
- Physical Description:
- 1 online resource (216 pages)
- Contained In:
- Dissertations Abstracts International 85-12B.
- Place of Publication:
- [Philadelphia, Pennsylvania] : University of Pennsylvania, 2022.
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- Language Note:
- English
- Summary:
- As machine learning models are deployed in a multitude of settings with increasing levels of influence and competency, there is growing interest in ensuring these models are robust and align with human intentions. To this end, we analyze robust models and adversarial inputs in a variety of settings. We explore statistical estimation under the adversarial setting of Wasserstein distribution shifts, where every data point may undergo a bounded perturbation. We analyze several statistical problems, including location estimation, linear regression, and non-parametric density estimation. Furthermore, we evaluate alignment in modern foundation models, and propose automated methods to construct adversarial inputs. We develop black-box automated algorithms to generate adversarial prompts for text-to-image models and jailbreaks for language models. Lastly, we introduce a benchmark, JailbreakBench, for reproducible jailbreak evaluation.
- Notes:
- Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
- Advisors: Dobriban, Edgar; Committee members: Altschuler, Jason; Hassani, Hamed; Wong, Eric.
- Department: Statistics and Data Science.
- Ph.D. University of Pennsylvania 2024.
- Local Notes:
- School code: 0175
- ISBN:
- 9798382830964
- Access Restriction:
- Restricted for use by site license.
The Penn Libraries is committed to describing library materials using current, accurate, and responsible language. If you discover outdated or inaccurate language, please fill out this feedback form to report it and suggest alternative language.