Learning to Optimize in Swarms
This work addresses the need for more robust and efficient optimization methods in machine learning and applications like protein docking, though it appears incremental as it builds on existing meta-optimizer frameworks.
The paper tackles the problem of meta-optimizers being limited to point-based and uncertainty-unaware algorithms by proposing a new meta-optimizer that learns in the space of both point-based and population-based algorithms, using LSTMs with attention and uncertainty measures, and demonstrates outperformance over existing competitors in non-convex test functions and protein-docking applications.
Learning to optimize has emerged as a powerful framework for various optimization and machine learning tasks. Current such "meta-optimizers" often learn in the space of continuous optimization algorithms that are point-based and uncertainty-unaware. To overcome the limitations, we propose a meta-optimizer that learns in the algorithmic space of both point-based and population-based optimization algorithms. The meta-optimizer targets at a meta-loss function consisting of both cumulative regret and entropy. Specifically, we learn and interpret the update formula through a population of LSTMs embedded with sample- and feature-level attentions. Meanwhile, we estimate the posterior directly over the global optimum and use an uncertainty measure to help guide the learning process. Empirical results over non-convex test functions and the protein-docking application demonstrate that this new meta-optimizer outperforms existing competitors.