# Independently Controllable Features

arXiv: https://arxiv.org/abs/1703.07718 · HTML: https://mlsanity.com/paper/1703.07718/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Mar 2017.

## Built on by (most cited first, 11 of 11)

- [Towards Causal Representation Learning](https://mlsanity.com/paper/2102.11107/) (2021)
- [Learning Neuro-Symbolic Skills for Bilevel Planning](https://mlsanity.com/paper/2206.10680/) (2022)
- [Learning Disentangled Representations in the Imaging Domain](https://mlsanity.com/paper/2108.12043/) (2021)
- [Causal models in string diagrams](https://mlsanity.com/paper/2304.07638/) (2023)
- [Interpretable Machine Learning in Physics: A Review](https://mlsanity.com/paper/2503.23616/) (2025)
- [Integrating Contrastive Learning with Dynamic Models for Reinforcement Learning from Images](https://mlsanity.com/paper/2203.01810/) (2022)
- [Towards a Grounded Theory of Causation for Embodied AI](https://mlsanity.com/paper/2206.13973/) (2022)
- [DisTop: Discovering a Topological representation to learn diverse and rewarding skills](https://mlsanity.com/paper/2106.03853/) (2021)
- [Temporal Abstractions-Augmented Temporally Contrastive Learning: An Alternative to the Laplacian in RL](https://mlsanity.com/paper/2203.11369/) (2022)
- [No Prior Mask: Eliminate Redundant Action for Deep Reinforcement Learning](https://mlsanity.com/paper/2312.06258/) (2023)
- [Meaning Versus Information, Prediction Versus Memory, and Question Versus Answer](https://mlsanity.com/paper/2107.13393/) (2021)

## Abstract (arXiv)

Finding features that disentangle the different causes of variation in real data is a difficult task, that has nonetheless received considerable attention in static domains like natural images. Interactive environments, in which an agent can deliberately take actions, offer an opportunity to tackle this task better, because the agent can experiment with different actions and observe their effects. We introduce the idea that in interactive environments, latent factors that control the variation in observed data can be identified by figuring out what the agent can control. We propose a naive method to find factors that explain or measure the effect of the actions of a learner, and test it in illustrative experiments.