# Unlearning Works Better Than You Think: Local Reinforcement-Based Selection of Auxiliary Objectives

arXiv: https://arxiv.org/abs/2504.14418 · HTML: https://mlsanity.com/paper/2504.14418/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods.
- Builds on 5 of its 31 references that are in the map.
- First posted on arXiv 19 Apr 2025.

## Placed in

- Deep reinforcement learning methods / [RL for games and combinatorial optimization](https://mlsanity.com/deeprl/rl-for-games-and-combinatorial-optimization/)

## Builds on (most cited first, 5 of 5)

- [Evolutionary Reinforcement Learning: A Survey](https://mlsanity.com/paper/2303.04150/) (2023)
- [Bridging Evolutionary Algorithms and Reinforcement Learning: A Comprehensive Survey on Hybrid Algorithms](https://mlsanity.com/paper/2401.11963/) (2024)
- [Reinforcement Learning-assisted Evolutionary Algorithm: A Survey and Research Opportunities](https://mlsanity.com/paper/2308.13420/) (2023)
- [An Extended Jump Functions Benchmark for the Analysis of Randomized Search Heuristics](https://mlsanity.com/paper/2105.03090/) (2021)
- [Hyper-Heuristics Can Profit From Global Variation Operators](https://mlsanity.com/paper/2407.14237/) (2024)

## Abstract (arXiv)

We introduce Local Reinforcement-Based Selection of Auxiliary Objectives (LRSAO), a novel approach that selects auxiliary objectives using reinforcement learning (RL) to support the optimization process of an evolutionary algorithm (EA) as in EA+RL framework and furthermore incorporates the ability to unlearn previously used objectives. By modifying the reward mechanism to penalize moves that do no increase the fitness value and relying on the local auxiliary objectives, LRSAO dynamically adapts its selection strategy to optimize performance according to the landscape and unlearn previous objectives when necessary. We analyze and evaluate LRSAO on the black-box complexity version of the non-monotonic Jump function, with gap parameter ℓ, where each auxiliary objective is beneficial at specific stages of optimization. The Jump function is hard to optimize for evolutionary-based algorithms and the best-known complexity for reinforcement-based selection on Jump was O(n2 log(n) / ℓ). Our approach improves over this result to achieve a complexity of Theta(n2 / ℓ2 + n log(n)) resulting in a significant improvement, which demonstrates the efficiency and adaptability of LRSAO, highlighting its potential to outperform traditional methods in complex optimization scenarios.