mlsanityData to 14 Oct 2026

Unsupervised Real-Time Control through Variational Empowerment

arXiv 1710.05101 ↗First posted 13 Oct 2017stat.MLcited by 18

Not on the map. Why?

Abstract (arXiv)

We introduce a methodology for efficiently computing a lower bound to empowerment, allowing it to be used as an unsupervised cost function for policy learning in real-time control. Empowerment, being the channel capacity between actions and states, maximises the influence of an agent on its near future. It has been shown to be a good model of biological behaviour in the absence of an extrinsic goal. But empowerment is also prohibitively hard to compute, especially in nonlinear continuous spaces. We introduce an efficient, amortised method for learning empowerment-maximising policies. We demonstrate that our algorithm can reliably handle continuous dynamical systems using system dynamics learned from raw data. The resulting policies consistently drive the agents into states where they can use their full potential.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). First posted on arXiv 13 Oct 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 4; 2022: 5; 2023: 2; 2024: 1; 2025: 1; 2026: 5.

18 papers.

Sort:

By area:

  1. 2022Emergent Bartering Behaviour in Multi-Agent Reinforcement Learningcited by 19
  2. 2021Variational Empowerment as Representation Learning for Goal-Based Reinforcement Learningcited by 15
  3. 2021Mind the Gap when Conditioning Amortised Inference in Sequential Latent-Variable Modelscited by 14
  4. 2021Information is Power: Intrinsic Control via Information Capturecited by 11
  5. 2021Understanding the Origin of Information-Seeking Exploration in Probabilistic Objectives for Controlcited by 10
  6. 2022Intrinsic Motivation in Dynamical Control Systemscited by 8
  7. 2025Information-Theoretic Policy Pre-Training with Empowermentcited by 2
  8. 2026Latent Matters: Learning Deep State-Space Modelscited by 2
  9. 2026Temporal Representations for Exploration: Learning Complex Exploratory Behavior without Extrinsic Rewardscited by 2
  10. 2023Filter-Aware Model-Predictive Controlcited by 1
  11. 2024Latent-Predictive Empowerment: Measuring Empowerment without a Simulatorcited by 1
  12. 2026Emergence of Physical Intelligence via Controllable Information Productioncited by 1
6 more
  1. 2022Reliably Re-Acting to Partner's Actions with the Social Intrinsic Motivation of Transfer Empowermentcited by 0
  2. 2022Ball-and-socket joint pose estimation using magnetic fieldcited by 0
  3. 2022PRISM: Probabilistic Real-Time Inference in Spatial World Modelscited by 0
  4. 2023Generative Intrinsic Optimization: Intrinsic Control with Model Learningcited by 0
  5. 2026Bellman Meets Lyapunov: Unsupervised Reinforcement Learning via Mastering Chaoscited by 0
  6. 2026Scaling Empowered Agentscited by 0