# Goal Recognition using Actor-Critic Optimization

arXiv: https://arxiv.org/abs/2501.01463 · HTML: https://mlsanity.com/paper/2501.01463/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods.
- First posted on arXiv 31 Dec 2024.

## Placed in

- Deep reinforcement learning methods / [Deep RL algorithms and representations](https://mlsanity.com/deeprl/deep-rl-algorithms-and-representations/)

## Built on by (most cited first, 5 of 5)

- [GRAIL: Goal Recognition Alignment through Imitation Learning](https://mlsanity.com/paper/2602.14252/) (2026)
- [General Dynamic Goal Recognition using Goal-Conditioned and Meta Reinforcement Learning](https://mlsanity.com/paper/2505.09737/) (2025)
- [GRAML: Goal Recognition As Metric Learning](https://mlsanity.com/paper/2505.03941/) (2025)
- [Online Dynamic Goal Recognition in Gym Environments](https://mlsanity.com/paper/2509.23244/) (2025)
- [IntentVLM: Open-Vocabulary Intention Recognition through Forward-Inverse Modeling with Video-Language Models](https://mlsanity.com/paper/2604.24002/) (2026)

## Abstract (arXiv)

Goal Recognition aims to infer an agent's goal from a sequence of observations. Existing approaches often rely on manually engineered domains and discrete representations. Deep Recognition using Actor-Critic Optimization (DRACO) is a novel approach based on deep reinforcement learning that overcomes these limitations by providing two key contributions. First, it is the first goal recognition algorithm that learns a set of policy networks from unstructured data and uses them for inference. Second, DRACO introduces new metrics for assessing goal hypotheses through continuous policy representations. DRACO achieves state-of-the-art performance for goal recognition in discrete settings while not using the structured inputs used by existing approaches. Moreover, it outperforms these approaches in more challenging, continuous settings at substantially reduced costs in both computing and memory. Together, these results showcase the robustness of the new algorithm, bridging traditional goal recognition and deep reinforcement learning.