mlsanityData to 7 Oct 2026

The Natural Language of Actions

arXiv 1902.01119 ↗First posted 4 Feb 2019cs.AI cs.CL cs.LGcited by 28

Not on the map. Why?

Abstract (arXiv)

We introduce Act2Vec, a general framework for learning context-based action representation for Reinforcement Learning. Representing actions in a vector space help reinforcement learning algorithms achieve better performance by grouping similar actions and utilizing relations between different actions. We show how prior knowledge of an environment can be extracted from demonstrations and injected into action vector representations that encode natural compatible behavior. We then use these for augmenting state representations as well as improving function approximation of Q-values. We visualize and test action embeddings in three domains including a drawing task, a high dimensional navigation task, and the large action space domain of StarCraft II.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). First posted on arXiv 4 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 6; 2023: 9; 2024: 2; 2025: 4; 2026: 1.

28 papers.

Sort:

By area:

  1. 2022Tutorial on amortized optimizationcited by 83
  2. 2023For SALE: State-Action Representation Learning for Deep Reinforcement Learningcited by 73
  3. 2022Supervised Training of Conditional Monge Mapscited by 60
  4. 2022Off-Policy Evaluation for Large Action Spaces via Embeddingscited by 53
  5. 2025On Entropy Control in LLM-RL Algorithmscited by 39
  6. 2023Off-Policy Evaluation for Large Action Spaces via Conjunct Effect Modelingcited by 34
  7. 2022Continuous MDP Homomorphisms and Homomorphic Policy Gradientcited by 24
  8. 2021Uncertainty Estimation Using Riemannian Model Dynamics for Offline Reinforcement Learningcited by 10
  9. 2022Generative Planning for Temporally Coordinated Exploration in Reinforcement Learningcited by 10
  10. 2021Action Redundancy in Reinforcement Learningcited by 9
  11. 2024Embedding-Aligned Language Modelscited by 9
  12. 2023Policy Gradient Methods in the Presence of Symmetries and State Abstractionscited by 8
16 more
  1. 2023Learning Action Embeddings for Off-Policy Evaluationcited by 7
  2. 2025Reinforcement Learning with Discrete Diffusion Policies for Combinatorial Action Spacescited by 7
  3. 2023No Prior Mask: Eliminate Redundant Action for Deep Reinforcement Learningcited by 6
  4. 2024Stochastic Q-learning for Large Discrete Action Spacescited by 5
  5. 2023Dynamic Neighborhood Construction for Structured Large Discrete Action Spacescited by 4
  6. 2023Representation-Driven Reinforcement Learningcited by 4
  7. 2025Learning Actionable World Models for Industrial Process Controlcited by 4
  8. 2021Locality Matters: A Scalable Value Decomposition Approach for Cooperative Multi-Agent Reinforcement Learningcited by 3
  9. 2021Learning Natural Language Generation from Scratchcited by 2
  10. 2022Reinforcement Learning with Large Action Spaces for Neural Machine Translationcited by 2
  11. 2023Diverse Policy Optimization for Structured Action Spacecited by 1
  12. 2021Learning Routines for Effective Off-Policy Reinforcement Learningcited by 0
  13. 2021More Efficient Exploration with Symbolic Priors on Action Sequence Equivalencescited by 0
  14. 2023DCT: Dual Channel Training of Action Embeddings for Reinforcement Learning with Large Discrete Action Spacescited by 0
  15. 2025LOD: Latent Objective Discovery in Heterogeneous Multi-Agent Reinforcement Learningcited by 0
  16. 2026LLM-Guided Contextual Action Evaluation for Operational Decisions in Industrial Processescited by 0