From federated learning to X-Learning: breaking the barriers of decentrality through random walks

Key Points

• This work aims to introduce and explore the concept of $${ extbf{X}}$$-Learning, a decentralized learning architecture.
• Discussed design considerations and degrees of freedom in $${ extbf{X}}$$-Learning.
• Explored connections between $${ extbf{X}}$$-Learning, graph theory, and Markov chains.
• Outlined open research directions for further investigation.
• Established relationships between decentralized concepts and graph theory.
• Highlighted the role of Markov chains in understanding $${ extbf{X}}$$-Learning.