Search papers, labs, and topics across Lattice.
This paper introduces a novel constrained Bayesian experimental design (BED) approach that combines offline pre-training of an amortized policy and a posterior network with online multi-step lookahead planning using scenario trees. This allows for adaptation to dynamic constraints like budget limitations and varying costs, which are often present in real-world tasks but not addressed by existing BED methods. The proposed method demonstrably yields more informative design sequences across various constrained BED tasks with minimal computational overhead.
Real-world experimental design gets a boost: this method adapts to budget limits and changing costs, outperforming existing approaches.
Bayesian experimental design (BED) is a principled framework for data-efficient design of sequential experiments. However, existing BED methods are unable to adapt to dynamic constraints inherent in real-world tasks due to budget limitations, varying costs, or physical constraints that restrict how designs evolve over time. In this paper, we introduce a novel approach to BED that enables constrained optimization of experimental designs by combining offline pre-training of an amortized policy and a posterior network with online multi-step lookahead planning using scenario trees. We empirically demonstrate that our method yields substantially more informative design sequences than existing methods across a range of constrained BED tasks, while incurring only a modest additional computational overhead.
