Technology Category
Machine Learning: Causal LearningReasoning under Uncertainty: CausalityKnowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
Causal Abstraction Inference under Lossy Representations
🤖 AI Summary
Existing causal abstraction frameworks struggle with lossy abstractions—i.e., many-to-one mappings—because the conventional “abstraction invariance” assumption fails when multiple low-level interventions map to a single high-level intervention. To address this, we propose the **projection abstraction framework**, which relaxes this assumption and establishes the first rigorous causal abstraction theory for lossy representations. Our approach constructs learnable projection mappings that preserve causal semantics when transforming from complex, low-dimensional models to concise, high-dimensional abstract models. We introduce a graph-structural identifiability criterion enabling high-level causal structure inference from finite observational data. Theoretically, we prove cross-level transferability of the framework for observational, interventional, and counterfactual queries. Empirical evaluation on high-dimensional image domains demonstrates accurate recovery of high-level causal structures, significantly enhancing both interpretability and computational efficiency in modeling complex systems. (149 words)
Technology Category
Application Category
📝 Abstract
The study of causal abstractions bridges two integral components of human intelligence: the ability to determine cause and effect, and the ability to interpret complex patterns into abstract concepts. Formally, causal abstraction frameworks define connections between complicated low-level causal models and simple high-level ones. One major limitation of most existing definitions is that they are not well-defined when considering lossy abstraction functions in which multiple low-level interventions can have different effects while mapping to the same high-level intervention (an assumption called the abstract invariance condition). In this paper, we introduce a new type of abstractions called projected abstractions that generalize existing definitions to accommodate lossy representations. We show how to construct a projected abstraction from the low-level model and how it translates equivalent observational, interventional, and counterfactual causal queries from low to high-level. Given that the true model is rarely available in practice we prove a new graphical criteria for identifying and estimating high-level causal queries from limited low-level data. Finally, we experimentally show the effectiveness of projected abstraction models in high-dimensional image settings.
Problem
Research questions and friction points this paper is trying to address.
Generalizing causal abstractions to accommodate lossy representation functions
Translating causal queries across observational interventional and counterfactual levels
Identifying estimable high-level causal effects from limited low-level data
Innovation
Methods, ideas, or system contributions that make the work stand out.
Projected abstractions handle lossy representations
Construct abstractions from low-level causal models
Graphical criteria identify high-level causal queries
🔎 Similar Papers
No similar papers found.
