aiqbench analysis
ReSCENE addresses a critical bottleneck in federated continual learning by shifting anti-forgetting mechanisms from resource-constrained clients to the server through condensed data surrogates. This approach significantly reduces client computational overhead and upload bandwidth while maintaining superior accuracy across standard benchmarks. The method offers a practical pathway for deploying large-scale, evolving AI models in distributed environments without sacrificing historical knowledge.
ReSCENE: Server-Side Replay for Structural Mitigation of Catastrophic Forgetting in Federated Continual Learning
arXiv:2609.38833v1 Announce Type: new Abstract: Federated continual learning must integrate new tasks over time without losing earlier-task knowledge. Most existing methods attach an anti-forgetting mechanism to the client-trained, server-aggregated loop of federated learning, which holds back new learning to preserve earlier knowledge and burdens resource-constrained clients. We propose ReSCENE, which structurally mitigates catastrophic forgetting by having each client upload a small condensed surrogate of its local data while the server keeps the surrogates of past tasks and trains the global model on them together with the current task surrogates. For efficient server memory, we introduce temporal herding, which selects the more recent surrogates from the pool accumulated over a task into a compressed buffer. Our study provides a theoretical analysis showing that this buffer can represent the original task data more closely than full accumulation of all surrogates. Across CIFAR-10, CIFAR-100, and TinyImageNet, ReSCENE achieves the strongest accuracy over seven baselines, by up to $31.1$ points of average accuracy, while requiring as little as $0.11\times$ of the client computation and up to $179\times$ less upload than the model-update baselines. ReSCENE further demonstrates its effectiveness when scaled to larger client populations and larger models while remaining efficient, which makes it a practical method for federated continual learning.
Abstract:Federated continual learning must integrate new tasks over time without losing earlier-task knowledge. Most existing methods attach an anti-forgetting mechanism to the client-trained, server-aggregated loop of federated learning, which holds back new learning to preserve earlier knowledge and burdens resource-constrained clients. We propose ReSCENE, which structurally mitigates catastrophic forgetting by having each client upload a small condensed surrogate of its local data while the server keeps the surrogates of past tasks and trains the global model on them together with the current task surrogates. For efficient server memory, we introduce temporal herding, which selects the more recent surrogates from the pool accumulated over a task into a compressed buffer. Our study provides a theoretical analysis showing that this buffer can represent the original task data more closely than full accumulation of all surrogates. Across CIFAR-10, CIFAR-100, and TinyImageNet, ReSCENE achieves the strongest accuracy over seven baselines, by up to $31.1$ points of average accuracy, while requiring as little as $0.11\times$ of the client computation and up to $179\times$ less upload than the model-update baselines. ReSCENE further demonstrates its effectiveness when scaled to larger client populations and larger models while remaining efficient, which makes it a practical method for federated continual learning.
Machine Learning (cs.LG)
arXiv:2609.38833 [cs.LG]
(or arXiv:2609.38833v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.38833 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 295 words
Related coverage
FedOGL: Combating Catastrophic Forgetting in Federated Open-World Multimodal Graph Learning
arXiv · Jul 31, 2026
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · Jun 12, 2026
HyperDCM: Dynamic Cluster Memory Replay in Hyperbolic Space for Continual Robotic Navigation Across Scenes
arXiv · Jul 21, 2026
How to apply to AI safety fellowships (and beyond)
LessWrong · Oct 2, 2026