Benchmarking MolDAIS

Recent works [1,2] are increasingly turning towards active encoding of molecular feature spaces. The motivation behind active encoding is that a priori encodings may not exhibit a smooth response to an arbitrary molecular property, reducing the performance of sample-efficient optimization algorithms, such as Bayesian optimization.
This project will focus on the algorithm proposed in [1], applied to the benchmarking datasets with real-world noise features. Additionally, we plan to extend these algorithms to the constrained and multi-objective benchmarking datasets.

References:

  1. Sorourifar, F., Banker, T., & Paulson, J. A. (2024). Accelerating Black-Box Molecular Property Optimization by Adaptively Learning Sparse Subspaces. arXiv [q-Bio.BM]. Retrieved from http://arxiv.org/abs/2401.01398

  2. Maus, N., Jones, H. T., Moore, J. S., Kusner, M. J., Bradshaw, J., & Gardner, J. R. (2023). Local Latent Space Bayesian Optimization over Structured Inputs. arXiv [Cs.LG]. Retrieved from http://arxiv.org/abs/2201.11872