Constraint-Aware Discrete Black-Box Optimization Using Tensor Decomposition

ICLR 2026 Conference Submission16278 Authors

Keywords: Black-Box Optimization, Constrained Optimization, Tensor Decomposition, Discrete Optimization, Surrogate Modeling, Sequential Model-Based Optimization

Abstract:

Discrete black-box optimization has been addressed using approaches such as Sequential Model-Based Optimization (SMBO), which aim to improve sample efficiency by fitting surrogate models that approximate a costly objective function over a discrete search space.
In many real-world problems, the set of feasible inputs such as valid parameter configurations in engineering design is often known in advance.
However, existing surrogate modeling techniques generally fail to capture feasibility constraints associated with such inputs.
In this paper, we propose a surrogate modeling approach based on tensor decomposition that captures the structure of discrete search spaces while directly integrating feasibility information.
To implement this approach, we formulate surrogate model training as a constrained polynomial optimization problem and solve a relaxed version of it.
Our experiments on both synthetic and real-world benchmarks, including a pressure vessel design task, demonstrate that the proposed method improves sample efficiency by effectively guiding the search away from infeasible regions.

Primary Area: optimization

Submission Number: 16278