Neural Genetic Search in Discrete Spaces
This work addresses the need for effective test-time search methods for deep generative models, which is significant for researchers and practitioners working with these models.
The authors tackled the problem of improving deep generative models' performance at test time and achieved effective results through their Neural Genetic Search method. The approach demonstrated its flexibility across three distinct domains.
Effective search methods are crucial for improving the performance of deep generative models at test time. In this paper, we introduce a novel test-time search method, Neural Genetic Search (NGS), which incorporates the evolutionary mechanism of genetic algorithms into the generation procedure of deep models. The core idea behind NGS is its crossover, which is defined as parent-conditioned generation using trained generative models. This approach offers a versatile and easy-to-implement search algorithm for deep generative models. We demonstrate the effectiveness and flexibility of NGS through experiments across three distinct domains: routing problems, adversarial prompt generation for language models, and molecular design.