C-HDNet: A Fast Hyperdimensional Computing Based Method for Causal Effect Estimation from Networked Observational Data
This addresses causal inference in networked data for researchers and practitioners, offering a simpler and faster alternative to computationally intensive methods, though it appears incremental as it builds on existing matching techniques with a new computational approach.
The paper tackled the problem of estimating causal effects from observational data with network confounding, where neighbors influence treatments and outcomes, and proposed a fast hyperdimensional computing-based matching method that outperforms or matches state-of-the-art methods, achieving similar error rates with roughly 10 times lower runtime.
We consider the problem of estimating causal effects from observational data in the presence of network confounding. In this context, an individual's treatment assignment and outcomes may be affected by their neighbors within the network. We propose a novel matching technique which leverages hyperdimensional computing to model network information and improve predictive performance. We present results of extensive experiments which show that the proposed method outperforms or is competitive with the state-of-the-art methods for causal effect estimation from network data, including advanced computationally demanding deep learning methods. Further, our technique benefits from simplicity and speed, with roughly an order of magnitude lower runtime compared to state-of-the-art methods, while offering similar causal effect estimation error rates.