Search papers, labs, and topics across Lattice.
This paper introduces BOBa, a bandit-guided surrogate optimization framework designed to enhance sample efficiency in the context of structure-based drug discovery by avoiding the computational burden of full-library surrogate inference. By treating partitions of the action space as multi-armed bandit arms, BOBa adaptively allocates computational resources to the most promising candidates, balancing exploration and exploitation effectively. Experimental results demonstrate that this approach significantly improves screening performance while reducing inference costs, paving the way for practical optimization in ultra-large molecular libraries.
BOBa's innovative bandit-guided approach cuts down computational costs while boosting screening efficiency in the vast chemical space of drug discovery.
Identifying high-utility candidates from massive discrete spaces under expensive evaluations is a recurring challenge across the sciences, with structure-based drug discovery as a prominent example. While surrogate-based optimization can increase sample efficiency by reducing the number of expensive evaluations, modern molecular libraries have reached billions to trillions of compounds, making full-library surrogate inference itself a major computational bottleneck. We introduce BOBa, a bandit-guided surrogate optimization framework that eliminates full-library inference by adaptively allocating computation across partitions of the action space. By treating partitions as arms in a multi-armed bandit, BOBa concentrates inference and evaluations on empirically promising partitions while maintaining principled exploration. Experiments on real-world synthesis-on-demand libraries demonstrate that optimism-under-uncertainty bandits, combined with meaningful action space partitioning, are essential for effective allocation of inference and evaluations. Our findings reveal a tunable tradeoff between screening performance and surrogate inference cost, which supports practical optimization over current libraries, and establishes a viable route to ultra-large library virtual screening.
