Timezone: »

FedHPO-Bench: A Benchmark Suite for Federated Hyperparameter Optimization
Zhen WANG · Weirui Kuang · Ce Zhang · Bolin Ding · Yaliang Li

Tue Jul 25 02:00 PM -- 04:30 PM (PDT) @ Exhibit Hall 1 #510

Research in the field of hyperparameter optimization (HPO) has been greatly accelerated by existing HPO benchmarks. Nonetheless, existing efforts in benchmarking all focus on HPO for traditional learning paradigms while ignoring federated learning (FL), a promising paradigm for collaboratively learning models from dispersed data. In this paper, we first identify some uniqueness of federated hyperparameter optimization (FedHPO) from various aspects, showing that existing HPO benchmarks no longer satisfy the need to study FedHPO methods. To facilitate the research of FedHPO, we propose and implement a benchmark suite FedHPO-Bench that incorporates comprehensive FedHPO problems, enables flexible customization of the function evaluations, and eases continuing extensions. We conduct extensive experiments based on FedHPO-Bench to provide the community with more insights into FedHPO. We open-sourced FedHPO-Bench at https://github.com/alibaba/FederatedScope/tree/master/benchmark/FedHPOBench.

Author Information

Zhen WANG (Alibaba Group)
Weirui Kuang (Alibaba Group)
Ce Zhang (ETH Zurich)
Bolin Ding (Alibaba Group)
Yaliang Li (Alibaba Group)

More from the Same Authors