RESIST: Resilient Decentralized Learning Using Consensus Gradient Descent
cs.LG, math.OC, stat.ML
Submitted: 2025-02-11
Updated: 2026-09-19
Comments: 99 pages, 14 figures, and 1 table; final version published in Transactions on Machine Learning Research (09/2026). Published version: https://openreview.net/forum?id=aNu8S9O6Ug
Journal ref: Transactions on Machine Learning Research, September 2026
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Byzantine Multi-Agent Optimization: Part II
- Byzantine-Robust Decentralized Learning via ClippedGossip
- Generalized Byzantine-tolerant SGD
- Phocas: dimensional Byzantine-resilient stochastic gradient descent
- Securing Distributed Gradient Descent in High Dimensional Statistical Learning
- Zeno: Distributed Stochastic Gradient Descent with Suspicion-based Fault-tolerance
- On the Geometric Convergence of Byzantine-Resilient Distributed Optimization Algorithms
- Matrix Representation of Iterative Approximate Byzantine Consensus in Directed Graphs
- Byzantine-robust Federated Learning through Collaborative Malicious Gradient Filtering
- SignSGD with Federated Defense: Harnessing Adversarial Attacks through Gradient Sign Decoding
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks