Generalizable Hyperparameter Optimization for Federated Learning on Non-IID Cancer Images
This addresses performance variability in federated learning for medical imaging, but it is incremental as it builds on existing methods with a heuristic approach.
The paper tackled the problem of hyperparameter optimization in federated learning for cancer histopathology under non-IID data, finding that a simple cross-dataset aggregation heuristic achieved competitive classification performance.
Deep learning for cancer histopathology training conflicts with privacy constraints in clinical settings. Federated Learning (FL) mitigates this by keeping data local; however, its performance depends on hyperparameter choices under non-independent and identically distributed (non-IID) client datasets. This paper examined whether hyperparameters optimized on one cancer imaging dataset generalized across non-IID federated scenarios. We considered binary histopathology tasks for ovarian and colorectal cancers. We perform centralized Bayesian hyperparameter optimization and transfer dataset-specific optima to the non-IID FL setup. The main contribution of this study is the introduction of a simple cross-dataset aggregation heuristic by combining configurations by averaging the learning rates and considering the modal optimizers and batch sizes. This combined configuration achieves a competitive classification performance.