Research map / Federated learning and privacy / Personalized and heterogeneous FL
Social Hippocampus Memory Learning
Liping Yi, Zhiming Zhao, Kewen Zhu, Xiang Li, Zhiwei Shang, Qinghua Hu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove all memory modules (variant D) (ablation) | SoHip · CIFAR-100 (N=100) | test accuracy (%): 61.08 -> 63.33 | Improved |
| remove all memory modules (variant D) (ablation) | SoHip · Tiny-ImageNet (N=100) | test accuracy (%): 43.52 -> 46.20 | Improved |
| add importance gating in short-term memory abstraction (ablation) | SoHip · CIFAR-100 and Tiny-ImageNet | test accuracy (%) | Improved |
| add hippocampus-inspired consolidation instead of replacing long-term memory with short-term memory (ablation) | SoHip · CIFAR-100 and Tiny-ImageNet | test accuracy (%) | Improved |
| add individual-collective memory fusion (ablation) | SoHip · CIFAR-100 and Tiny-ImageNet | test accuracy (%) | Improved |
| increase memory dimension m (ablation) | SoHip · CIFAR-100 and Tiny-ImageNet | test accuracy (%) | Mixed |
| reduce non-IID degree (more classes per agent or larger Dirichlet alpha) (ablation) | SoHip · CIFAR-100 and Tiny-ImageNet | test accuracy (%) | Hurt |
| increase number of agents from 100 to 300 (ablation) | SoHip · CIFAR-100 | average test accuracy (%): 63.33 -> 50.10 | Hurt |
| share memory instead of model parameters or auxiliary model (SoHip vs FedMRL) | heterogeneous FL · Tiny-ImageNet (N=100) | average test accuracy (%): 37.42 -> 46.20 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- pFedES — pFedES: Generalized Proxy Feature Extractor Sharing for Model Heterogeneous Personalized Federated Learning (AAAI 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.82)
- Federated Class-Incremental Learning: A Hybrid Approach Using Latent Exemplars and Data-Free Techniques to Address Local and Global Forgetting (ICLR 2025 · similarity 0.82)
- G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems (NeurIPS 2025 · similarity 0.82)
- Socialized Coevolution: Advancing a Better World through Cross-Task Collaboration (ICML 2025 · similarity 0.82)
- HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models (NeurIPS 2024 · similarity 0.82)
- Socialized Learning: Making Each Other Better Through Multi-Agent Collaboration (ICML 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
