<ul><li>A generative, end-to-end solver for black-box combinatorial optimization has been proposed.</li><li>Inspired by annealing-based algorithms, a neural network is trained to model the Boltzmann distribution of the black-box objective.</li><li>The network's conditioning on temperature allows capturing a range of distributions, aiding in global optimization and improving sample efficiency.</li><li>The approach shows competitive performance on challenging combinatorial tasks with limited or unlimited query budgets.</li></ul>

A Generative Neural Annealer for Black-Box Combinatorial Optimization

Discover more