Towards Intervention-centric Causal Reasoning In Learning Agents

Abstract

Interventions are central to causal learning and reasoning. Yet ultimately an intervention is an abstraction: an agent embedded in a physical environment (perhaps modeled as a Markov decision process) does not typically come equipped with the notion of an intervention -- its action space is typically ego-centric, without actions of the form `intervene on X'. Such a correspondence between ego-centric actions and interventions would be challenging to hard-code. It would instead be better if an agent learnt which sequence of actions allow it to make targeted manipulations of the environment, and learnt corresponding representations that permitted learning from observation. Here we show how a meta-learning approach can be used to perform causal learning in this challenging setting, where the action-space is not a set of interventions and the observation space is a high-dimensional space with a latent causal structure. A meta-reinforcement learning algorithm is used to learn relationships t

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keylansdell2020towards

Related papers

• Resolving Spurious Correlations In Causal Models Of Environments Via Interventions (2020)0.00
• Learning By Doing: An Online Causal Reinforcement Learning Framework With Causal-aware Policy (2024)1.56
• Causal Reinforcement Learning Using Observational And Interventional Data (2021)0.00
• Disentangled Representations For Causal Cognition (2024)4.52
• Reinforcement Learning Of Causal Variables Using Mediation Analysis (2020)5.24
• Explainable Reinforcement Learning Through A Causal Lens (2019)16.69
• Active Causal Experimentalist (ACE): Learning Intervention Strategies Via Direct Preference Optimization (2026)0.00
• Empowerment Gain And Causal Model Construction: Children And Adults Are Sensitive To Controllability And Variability In Their Causal Interventions (2025)0.00