# Algorithms for learning value-aligned policies considering admissibility relaxation

arXiv: https://arxiv.org/abs/2406.04838 · HTML: https://mlsanity.com/paper/2406.04838/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 5 of its 25 references that are in the map.
- First posted on arXiv 7 Jun 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

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

- [OpenAI Gym](https://mlsanity.com/paper/1606.01540/) (2016)
- [Reward Constrained Policy Optimization](https://mlsanity.com/paper/1805.11074/) (2018)
- [Safe Exploration in Continuous Action Spaces](https://mlsanity.com/paper/1801.08757/) (2018)
- [Value alignment: a formal approach](https://mlsanity.com/paper/2110.09240/) (2021)
- [Value Engineering for Autonomous Agents](https://mlsanity.com/paper/2302.08759/) (2023)

## Built on by (most cited first, 1 of 1)

- [Learning the Value Systems of Agents with Preference-based and Inverse Reinforcement Learning](https://mlsanity.com/paper/2602.04518/) (2026)

## Abstract (arXiv)

The emerging field of value awareness engineering claims that software agents and systems should be value-aware, i.e. they must make decisions in accordance with human values. In this context, such agents must be capable of explicitly reasoning as to how far different courses of action are aligned with these values. For this purpose, values are often modelled as preferences over states or actions, which are then aggregated to determine the sequences of actions that are maximally aligned with a certain value. Recently, additional value admissibility constraints at this level have been considered as well. However, often relaxed versions of these constraints are needed, and this increases considerably the complexity of computing value-aligned policies. To obtain efficient algorithms that make value-aligned decisions considering admissibility relaxation, we propose the use of learning techniques, in particular, we have used constrained reinforcement learning algorithms. In this paper, we present two algorithms, ε-ADQL for strategies based on local alignment and its extension ε-CADQL for a sequence of decisions. We have validated their efficiency in a water distribution problem in a drought scenario.