Explicit Gradient Learning
This addresses the limitation of classical BBO methods in real-world AI tasks by improving performance in high-dimensional, ill-behaved optimization problems.
The paper tackles the challenge of optimizing high-dimensional non-convex functions in Black-Box Optimization (BBO) by proposing Explicit Gradient Learning (EGL), a method that trains a neural network to directly estimate gradients, achieving state-of-the-art results on the COCO test suite and a high-dimensional image generation task.
Black-Box Optimization (BBO) methods can find optimal policies for systems that interact with complex environments with no analytical representation. As such, they are of interest in many Artificial Intelligence (AI) domains. Yet classical BBO methods fall short in high-dimensional non-convex problems. They are thus often overlooked in real-world AI tasks. Here we present a BBO method, termed Explicit Gradient Learning (EGL), that is designed to optimize high-dimensional ill-behaved functions. We derive EGL by finding weak-spots in methods that fit the objective function with a parametric Neural Network (NN) model and obtain the gradient signal by calculating the parametric gradient. Instead of fitting the function, EGL trains a NN to estimate the objective gradient directly. We prove the convergence of EGL in convex optimization and its robustness in the optimization of integrable functions. We evaluate EGL and achieve state-of-the-art results in two challenging problems: (1) the COCO test suite against an assortment of standard BBO methods; and (2) in a high-dimensional non-convex image generation task.