Unravelling in Collaborative Learning

Authors: Aymeric Capitaine, Etienne Boursier, Antoine Scheid, Eric Moulines, Michael Jordan, El-Mahdi El-Mhamdi, Alain Durmus

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We provide a rigorous framework for analyzing collaborative learning with strategic agents having data distributions of varying quality. ... We show that when data quality is private, a naive aggregation strategy... results in a complete unravelling. ... We present solutions to unravelling. ... Theorem 1. Assume H1, H2, H3. Problem (8) admits a unique solution... Theorem 2 (Unravelling). Assume H1, H2, H3, and H4. Let E S J be the set of pure-strategy Nash equilibria... Theorem 3. Assume H1, H2, H3, H4 and H5. B = (1, . . . , 1)T is a Nash equilibrium under ˆΓ with probability 1 δ.
Researcher Affiliation Academia Aymeric Capitaine1 Etienne Boursier2 Antoine Scheid1 Eric Moulines1 Michael I. Jordan3 El-Mahdi El-Mhamdi1 Alain Durmus1 1 Centre de Mathématiques Appliquées CNRS École polytechnique Palaiseau, 91120, France 2 INRIA Saclay, Université Paris Saclay, LMO Orsay, 91400, France 3 Inria, Ecole Normale Supérieure, PSL Research University Paris, 75, France
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks. Methods are described in prose and mathematical notation.
Open Source Code No The paper does not contain any explicit statement or link providing concrete access to source code for the methodology described.
Open Datasets No The paper describes a theoretical framework and does not conduct experiments on specific public or open datasets. The concept of 'samples' refers to theoretical data generation within the model.
Dataset Splits No The paper is theoretical and does not involve empirical validation with dataset splits for training, validation, or testing.
Hardware Specification No The paper does not describe any experiments or their computational execution, thus no hardware specifications are provided.
Software Dependencies No The paper does not describe any experiments or their computational execution, thus no specific software dependencies with version numbers are provided.
Experiment Setup No The paper is theoretical and does not contain details about an experimental setup, hyperparameters, or system-level training settings.