Summary of Fullcert: Deterministic End-to-end Certification For Training and Inference Of Neural Networks, by Tobias Lorenz et al.
FullCert: Deterministic End-to-End Certification for Training and Inference of Neural Networks
by Tobias Lorenz, Marta Kwiatkowska, Mario Fritz
First submitted to arxiv on: 17 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
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 This paper presents FullCert, an end-to-end certifier that provides sound, deterministic bounds for machine learning models against both training-time and inference-time attacks. The authors develop a novel certification paradigm that bounds all possible perturbations in the training data, which affects model parameters and predictions. This approach is achieved through a new open-source library called BoundFlow, enabling model training on bounded datasets. Experiments demonstrate FullCert’s feasibility on two datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Machine learning models can be vulnerable to attacks that manipulate training or inference data. To address this issue, the paper introduces FullCert, an innovative certifier that provides guarantees against both types of attacks. By bounding possible perturbations in training data and their impact on model parameters and predictions, FullCert offers robustness assurances. A new library called BoundFlow facilitates this approach by allowing model training on bounded datasets. |
Keywords
* Artificial intelligence * Inference * Machine learning
