Soft-Label Caching and Sharpening for Communication-Efficient Federated Distillation

We propose SCARLET, a communication-efficient distillation-based Federated Learning framework. By caching soft-labels to minimize redundant transmission and utilizing an enhanced Entropy Reduction Aggregation, it achieves up to 50% communication cost reduction while maintaining competitive accuracy.

Beyond Multiprocessing: A Real-World ML Workload Speedup with Python 3.13+ Free-Threading

研究室サーバーとKubeflowで実践するNotebook as a Service

rootlessコンテナのすゝめ - 研究室サーバーでもできる安全なコンテナ管理