GPU Accelerated Exhaustive Search for Optimal Ensemble of Black-Box Optimization Algorithms
This work provides a faster method for machine learning practitioners to find optimal ensembles of black-box optimization algorithms, which is an incremental improvement for hyperparameter tuning.
This paper demonstrates that a simple ensemble of black-box optimization algorithms can outperform individual optimizers. They developed a multi-GPU framework to accelerate the exhaustive search for such optimal ensembles, reducing search time from over 10 days on CPUs to under 24 hours on 8 GPUs, and this approach secured 2nd place in the NeurIPS 2020 black-box optimization challenge.
Black-box optimization is essential for tuning complex machine learning algorithms which are easier to experiment with than to understand. In this paper, we show that a simple ensemble of black-box optimization algorithms can outperform any single one of them. However, searching for such an optimal ensemble requires a large number of experiments. We propose a Multi-GPU-optimized framework to accelerate a brute force search for the optimal ensemble of black-box optimization algorithms by running many experiments in parallel. The lightweight optimizations are performed by CPU while expensive model training and evaluations are assigned to GPUs. We evaluate 15 optimizers by training 2.7 million models and running 541,440 optimizations. On a DGX-1, the search time is reduced from more than 10 days on two 20-core CPUs to less than 24 hours on 8-GPUs. With the optimal ensemble found by GPU-accelerated exhaustive search, we won the 2nd place of NeurIPS 2020 black-box optimization challenge.