Counterfactual Dynamics Forecasting – a New Setting of Quantitative Reasoning
DOI:
https://doi.org/10.1609/aaai.v37i2.25265Keywords:
CV: Visual Reasoning & Symbolic Representations, CV: Applications, CV: Interpretability and Transparency, CV: Low Level & Physics-Based Vision, CV: Representation Learning for Vision, ML: Causal Learning, ML: Deep Neural Architectures, ML: Deep Neural Network AlgorithmsAbstract
Rethinking and introspection are important elements of human intelligence. To mimic these capabilities, counterfactual reasoning has attracted attention of AI researchers recently, which aims to forecast the alternative outcomes for hypothetical scenarios (“what-if”). However, most existing approaches focused on qualitative reasoning (e.g., casual-effect relationship). It lacks a well-defined description of the differences between counterfactuals and facts, as well as how these differences evolve over time. This paper defines a new problem formulation - counterfactual dynamics forecasting - which is described in middle-level abstraction under the structural causal models (SCM) framework and derived as ordinary differential equations (ODEs) as low-level quantitative computation. Based on it, we propose a method to infer counterfactual dynamics considering the factual dynamics as demonstration. Moreover, the evolution of differences between facts and counterfactuals are modelled by an explicit temporal component. The experimental results on two dynamical systems demonstrate the effectiveness of the proposed method.Downloads
Published
2023-06-26
How to Cite
Liu, Y., Sun, Y., & Lim, J.-H. (2023). Counterfactual Dynamics Forecasting – a New Setting of Quantitative Reasoning. Proceedings of the AAAI Conference on Artificial Intelligence, 37(2), 1764-1771. https://doi.org/10.1609/aaai.v37i2.25265
Issue
Section
AAAI Technical Track on Computer Vision II