Metraipolis › archive › dc_bd153f0ee756
MAEM: Mitigating Recency Bias via a Replay-Free, Capacity-Controlled Dynamic Mixture of Experts for Class-Incremental Graph Learning — arXiv:2610.20238
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.20238
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T14:19:48.590Z
- SHA-256 of the archived text
a5c401259236b25e74c70149905cecb91a1fc903370dc718027f0677b258b8af- Stable URL of this page
https://metraipolis.com/documento/dc_bd153f0ee756
How to cite
MAEM: Mitigating Recency Bias via a Replay-Free, Capacity-Controlled Dynamic Mixture of Experts for Class-Incremental Graph Learning — arXiv:2610.20238 — arXiv (en), https://arxiv.org/abs/2610.20238 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bd153f0ee756 (Metraipolis archive dc_bd153f0ee756, sha256 a5c401259236b25e74c70149905cecb91a1fc903370dc718027f0677b258b8af)
BibTeX
@misc{dcbd153f0ee756,
title = {MAEM: Mitigating Recency Bias via a Replay-Free, Capacity-Controlled Dynamic Mixture of Experts for Class-Incremental Graph Learning — arXiv:2610.20238},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.20238},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bd153f0ee756; sha256 a5c401259236b25e74c70149905cecb91a1fc903370dc718027f0677b258b8af},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20238. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Graph Neural Networks (GNNs) excel on graph-structured data but suffer catastrophic forgetting when classes arrive sequentially. In class-incremental graph learning, predictions are often dominated by recently learned classes, causing severe recency bias. Dynamic expansion can isolate knowledge across sessions, but many methods assume task-incremental settings and require task identities at test time to select branches or heads---information unavailable in class-incremental learning. Learnable routers offer an alternative, yet continual fine-tuning the router on new data can bias routing toward recent components. We propose MAEM, a replay-free, capacity-controlled dynamic mixture-of-experts framework for class-incremental graph learning. MAEM builds representations with Simplified Graph Convolution (SGC) and trains a dedicated expert for each new session, then freezes it to protect prior knowledge. For routing without task IDs or a shared trainable router, each expert uses a masked autoencoder (MAE); test nodes are assigned to the expert with the lowest reconstruction error, reducing routing-level recency bias. When the expert pool reaches a fixed capacity, MAEM merges the least-used experts via Git Re-Basin alignment and further distills them into a fused expert using topology-aware pseudo data synthesized from MAE distributions (DeepInversion-style), without storing or revisiting historical graph data. On four class-incremental graph benchmarks, MAEM mitigates recency bias, outperforms strong continual graph learning baselines, and keeps model capacity bounded under strict replay-free learning. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML). Pagina del lavoro: https://arxiv.org/abs/2610.20238 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T13:53:35.712Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).