Copertina

Tighter Risk Bounds for Mixtures of Experts

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Akretche, Wissam, LeBlanc, Frédéric, Marchand, Mario
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Machine Learning Cryptography and Security
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Private Gradient Descent for Linear Regression: Tighter Error Bounds and Instance-Specific Uncertainty Estimation
di: Brown, Gavin, et al.
Pubblicazione: (2024)

Buffer Overflow in Mixture of Experts
di: Hayes, Jamie, et al.
Pubblicazione: (2024)

Differentially Private Training of Mixture of Experts Models
di: Tholoniat, Pierre, et al.
Pubblicazione: (2024)

Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model
di: Cebere, Tudor, et al.
Pubblicazione: (2024)

Enhancing the "Immunity" of Mixture-of-Experts Networks for Adversarial Defense
di: Han, Qiao, et al.
Pubblicazione: (2024)

Adversarial Sample-Based Approach for Tighter Privacy Auditing in Final Model-Only Scenarios
di: Yoon, Sangyeon, et al.
Pubblicazione: (2024)

Differentially Private Stochastic Gradient Descent with Fixed-Size Minibatches: Tighter RDP Guarantees with or without Replacement
di: Birrell, Jeremiah, et al.
Pubblicazione: (2024)

MERLOT: A Distilled LLM-based Mixture-of-Experts Framework for Scalable Encrypted Traffic Classification
di: Chen, Yuxuan, et al.
Pubblicazione: (2024)

RASA: Routing-Aware Safety Alignment for Mixture-of-Experts Models
di: Liang, Jiacheng, et al.
Pubblicazione: (2026)

Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs
di: Jiang, Yukun, et al.
Pubblicazione: (2026)

Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Data
di: Zhou, Yvonne, et al.
Pubblicazione: (2024)

PM-MOE: Mixture of Experts on Private Model Parameters for Personalized Federated Learning
di: Feng, Yu, et al.
Pubblicazione: (2025)

MoJE: Mixture of Jailbreak Experts, Naive Tabular Classifiers as Guard for Prompt Attacks
di: Cornacchia, Giandomenico, et al.
Pubblicazione: (2024)

Mixture of Robust Experts (MoRE):A Robust Denoising Method towards multiple perturbations
di: Cheng, Hao, et al.
Pubblicazione: (2021)

The geometry of invariant learning: an information-theoretic analysis of data augmentation and generalization
di: Bouyahia, Abdelali, et al.
Pubblicazione: (2026)

Stealing User Prompts from Mixture of Experts
di: Yona, Itay, et al.
Pubblicazione: (2024)

LLM Watermarking Using Mixtures and Statistical-to-Computational Gaps
di: Abdalla, Pedro, et al.
Pubblicazione: (2025)

Preserving Expert-Level Privacy in Offline Reinforcement Learning
di: Sharma, Navodita, et al.
Pubblicazione: (2024)

The Power of Sampling: Dimension-free Risk Bounds in Private ERM
di: Lee, Yin Tat, et al.
Pubblicazione: (2021)

Skellam Mixture Mechanism: a Novel Approach to Federated Learning with Differential Privacy
di: Bao, Ergute, et al.
Pubblicazione: (2022)

Privacy Amplification for the Gaussian Mechanism via Bounded Support
di: Hu, Shengyuan, et al.
Pubblicazione: (2024)

Tight Group-Level DP Guarantees for DP-SGD with Sampling via Mixture of Gaussians Mechanisms
di: Ganesh, Arun
Pubblicazione: (2024)

Federated Online Prediction from Experts with Differential Privacy: Separations and Regret Speed-ups
di: Gao, Fengyu, et al.
Pubblicazione: (2024)

Lower Bounds for Public-Private Learning under Distribution Shift
di: Setlur, Amrith, et al.
Pubblicazione: (2025)

Generalization Properties of Adversarial Training for $\ell_0$-Bounded Adversarial Attacks
di: Delgosha, Payam, et al.
Pubblicazione: (2024)

Certified Robust Accuracy of Neural Networks Are Bounded due to Bayes Errors
di: Zhang, Ruihan, et al.
Pubblicazione: (2024)

Closed-Form Bounds for DP-SGD against Record-level Inference
di: Cherubin, Giovanni, et al.
Pubblicazione: (2024)

On the Privacy Risk of In-context Learning
di: Duan, Haonan, et al.
Pubblicazione: (2024)

Detecting Financial Fraud with Hybrid Deep Learning: A Mix-of-Experts Approach to Sequential and Anomalous Patterns
di: Vallarino, Diego
Pubblicazione: (2025)

Lower Bounds for Differential Privacy Under Continual Observation and Online Threshold Queries
di: Cohen, Edith, et al.
Pubblicazione: (2024)

On the Growth of Mistakes in Differentially Private Online Learning: A Lower Bound Perspective
di: Dmitriev, Daniil, et al.
Pubblicazione: (2024)

Fine Grained Insider Risk Detection
di: Huber, Birkett, et al.
Pubblicazione: (2024)

An Improved Privacy and Utility Analysis of Differentially Private SGD with Bounded Domain and Smooth Losses
di: Liang, Hao, et al.
Pubblicazione: (2025)

The poison of dimensionality
di: Hoang, Lê-Nguyên
Pubblicazione: (2024)

Randomization Techniques to Mitigate the Risk of Copyright Infringement
di: Chen, Wei-Ning, et al.
Pubblicazione: (2024)

Capability and Robustness Cannot Both Be Free: An Information-Theoretic Bound for Vision-Language-Action Models
di: Tai, Jianwei
Pubblicazione: (2026)

From Mean to Extreme: Formal Differential Privacy Bounds on the Success of Real-World Data Reconstruction Attacks
di: Riess, Anneliese, et al.
Pubblicazione: (2024)

Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD
di: Kalinin, Nikita P., et al.
Pubblicazione: (2025)

Trade-off Functions for DP-SGD with Subsampling based on Random Shuffling: Tight Upper and Lower Bounds
di: van Dijk, Marten, et al.
Pubblicazione: (2026)

Mitigating Downstream Model Risks via Model Provenance
di: Wang, Keyu, et al.
Pubblicazione: (2024)