arXiv:2406.07709
공식 코드 · Papers with Code
Diagnosing and fixing common problems in Bayesian optimization for molecule design
원문 초록
Bayesian optimization (BO) is a principled approach to molecular design tasks. In this paper we explain three pitfalls of BO which can cause poor empirical performance: an incorrect prior width, over-smoothing, and inadequate acquisition function maximization. We show that with these issues addressed, even a basic BO setup is able to achieve the highest overall performance on the PMO benchmark for molecule design (Gao et al 2022). These results suggest that BO may benefit from more attention in the machine learning for molecules community.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
LLINBO: Trustworthy LLM-in-the-Loop Bayesian Optimization
Reinforcement Learning Increases Wind Farm Power Production by Enabling Closed-Loop Collaborative Control
Fast Bayesian Optimization of Function Networks with Partial Evaluations
Amortized variational transdimensional inference
VirnyFlow: A Design Space for Responsible Model Development
Bridging Quantum and Classical Computing in Drug Design: Architecture Principles for Improved Molecule Generation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.