B3O: Scalable Boltzmann Batch Bayesian Optimization
For practitioners needing large-batch Bayesian optimization, B3O offers a scalable and theoretically grounded method that avoids computational bottlenecks and approximation errors.
B3O reframes batch generation in Bayesian optimization as sampling from the Boltzmann distribution of the acquisition function, enabling scalable large-batch BO. It outperforms existing methods on synthetic benchmarks and real-world tasks like electrode design and race car configuration.
Modern engineering workflows increasingly rely on massive parallel simulation, driving the need for scalable, large-batch Bayesian Optimization (BO). Existing batch BO methods, however, incur large computational cost or rely on approximations that erode batch diversity. We propose B3O (Boltzmann Batch Bayesian Optimization), a framework that reframes batch generation as a pure sampling problem: drawing samples directly from the Boltzmann distribution defined by the acquisition function avoids the bottlenecks of existing large-batch methods. Theoretically, we prove that queries sampled from this distribution incur only negligible additional regret. Empirically, B3O outperforms existing batch BO methods on standard synthetic benchmarks and adapts robustly across complex applied tasks, including multi-objective electrode design and mixed-variable race car configuration.