Research map / Adversarial robustness and security / Attacks on text and graphs
The Double-Edged Nature of the Rashomon Set for Trustworthy Machine Learning
Ethan Hsu, Harry Chen, Chudi Zhong, Lesia Semenova
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| select farthest-in-Hamming-distance trees instead of closest | Rashomon tree release · DRAFT reconstruction attack | reconstruction error | Hurt |
| increase number of released diverse trees | Rashomon tree ensembles · COMPAS and other datasets | adversarial accuracy | Improved |
| increase number of released diverse trees | Rashomon tree ensembles · COMPAS and other datasets | reconstruction error | Hurt |
| increase distance from the optimal tree | Rashomon set trees · adversarial dataset | adversarial accuracy | Improved |
| modify 6% of dataset and recompute Rashomon set | Rashomon set · four datasets | model overlap: more than 80% of models remain | No clear effect |
| modify 2% of dataset and recompute Rashomon set | Rashomon set · four datasets | model overlap: nearly identical | No clear effect |
| increase attack strength to eta=l+I-1 on a single rule list | single rule list · binary datasets | adversarial loss gap | Hurt |
Closest papers by meaning
- Provable Robustness against Backdoor Attacks via the Primal-Dual Perspective on Differential Privacy (NeurIPS 2026 · similarity 0.86)
- ElliCE: Efficient and Provably Robust Algorithmic Recourse via the Rashomon Sets (NeurIPS 2025 · similarity 0.85)
- Reconstruction Attacks on Machine Unlearning: Simple Models are Vulnerable (NeurIPS 2024 · similarity 0.85)
- Trustworthy Machine Learning through Data-Specific Indistinguishability (ICML 2025 · similarity 0.85)
- Towards Trustworthy Federated Learning with Untrusted Participants (ICML 2025 · similarity 0.85)
- Certifiably Robust Model Evaluation in Federated Learning under Meta-Distributional Shifts (ICML 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
