# A gray-box approach for curriculum learning

## TL;DR

This paper introduces a gray-box formulation for curriculum learning in deep reinforcement learning, providing a structured approach with efficient numerical methods and promising initial results on benchmark tasks.

## Contribution

It proposes a novel gray-box reformulation of curriculum learning, along with efficient numerical methods to solve it, advancing beyond heuristic solutions.

## Key findings

- Preliminary results show the approach's viability.
- The method effectively addresses the curriculum scheduling problem.
- Initial benchmarks indicate promising performance improvements.

## Abstract

Curriculum learning is often employed in deep reinforcement learning to let the agent progress more quickly towards better behaviors. Numerical methods for curriculum learning in the literature provides only initial heuristic solutions, with little to no guarantee on their quality. We define a new gray-box function that, including a suitable scheduling problem, can be effectively used to reformulate the curriculum learning problem. We propose different efficient numerical methods to address this gray-box reformulation. Preliminary numerical results on a benchmark task in the curriculum learning literature show the viability of the proposed approach.