Abstract:Molecular property optimization (MPO) problems are inherently challenging since they are formulated over discrete, unstructured spaces and the labeling process involves expensive simulations or experiments, which fundamentally limits the amount of available data. Bayesian optimization (BO) is a powerful and popular framework for efficient optimization of noisy, black-box objective functions (e.g., measured property values), thus is a potentially attractive framework for MPO. To apply BO to MPO problems, one must select a structured molecular representation that enables construction of a probabilistic surrogate model. Many molecular representations have been developed, however, they are all high-dimensional, which introduces important challenges in the BO process -- mainly because the curse of dimensionality makes it difficult to define and perform inference over a suitable class of surrogate models. This challenge has been recently addressed by learning a lower-dimensional encoding of a SMILE or graph representation of a molecule in an unsupervised manner and then performing BO in the encoded space. In this work, we show that such methods have a tendency to "get stuck," which we hypothesize occurs since the mapping from the encoded space to property values is not necessarily well-modeled by a Gaussian process. We argue for an alternative approach that combines numerical molecular descriptors with a sparse axis-aligned Gaussian process model, which is capable of rapidly identifying sparse subspaces that are most relevant to modeling the unknown property function. We demonstrate that our proposed method substantially outperforms existing MPO methods on a variety of benchmark and real-world problems. Specifically, we show that our method can routinely find near-optimal molecules out of a set of more than $>100$k alternatives within 100 or fewer expensive queries.

On Accelerating Substrate Optimization Using Computational Gibbs Energy Barriers: A Numerical Consideration Utilizing a Computational Data Set

Kinetic Parameter Estimation of the Polyethylene Process by Bayesian Optimization

Monte-Carlo-simulation-based Optimization for Copolymerization Processes with Embedded Chemical Composition Distribution.

Dynamic Optimization of Batch Free Radical Polymerization with Conditional Modeling Formulation Through the Adaptive Smoothing Strategy

Programming Substrate-Independent Kinetic Barriers with Thermodynamic Binding Networks

Application of Genetic Algorithm to Estimating Kinetic Model Parameters of Propylene Ammoxidation

Comparative Study on Adaptive Bayesian Optimization for Batch Cooling Crystallization for Slow and Fast Kinetic Regimes

Incorporation of physical bounds on rate parameters for reaction mechanism optimization using genetic algorithms

Practical Path-based Bayesian Optimization

Determination of the Free Energies of Mixing of Organic Solutions Through a Combined Molecular Dynamics and Bayesian Statistics Approach.

CAPBO: A Cost‐aware Parallelized Bayesian Optimization Method for Chemical Reaction Optimization

Process-constrained batch Bayesian approaches for yield optimization in multi-reactor systems

Computed versus experimental energy barriers in solution: Influence of the type of the density functional approximation

Bayesian optimization for chemical products and functional materials

Accelerating Black-Box Molecular Property Optimization by Adaptively Learning Sparse Subspaces

Race to the bottom: Bayesian optimisation for chemical problems

Adaptive Catalyst Discovery Using Multicriteria Bayesian Optimization with Representation Learning

Extrapolative Bayesian Optimization with Gaussian Process and Neural Network Ensemble Surrogate Models

Optimization of heterogeneous continuous flow hydrogenation using FTIR inline analysis: a comparative study of multi-objective Bayesian optimization and kinetic modeling

Optimization of Fischer–Tropsch microchannel reactor using computational fluid dynamics and enveloped Bayesian optimization

Computed vs Experimental Energy Barriers in Solution: Influence of the type of the density functional approximation.