Samothrakis, Spyridon and Matran-Fernandez, Ana and Abdullahi, Umar and Fairbank, Michael and Fasli, Maria (2022) Grokking-like effects in counterfactual inference. In: 2022 International Joint Conference on Neural Networks (IJCNN), 2022-07-18 - 2022-07-23, Padua, Italy.
Samothrakis, Spyridon and Matran-Fernandez, Ana and Abdullahi, Umar and Fairbank, Michael and Fasli, Maria (2022) Grokking-like effects in counterfactual inference. In: 2022 International Joint Conference on Neural Networks (IJCNN), 2022-07-18 - 2022-07-23, Padua, Italy.
Samothrakis, Spyridon and Matran-Fernandez, Ana and Abdullahi, Umar and Fairbank, Michael and Fasli, Maria (2022) Grokking-like effects in counterfactual inference. In: 2022 International Joint Conference on Neural Networks (IJCNN), 2022-07-18 - 2022-07-23, Padua, Italy.
Abstract
We show that a typical neural network, which ignores any covariate/feature re-balancing, can be as effective as any explicit counterfactual method. We adopt the architecture of TARNet—a simple neural network with two heads (one for treatment, one for control) which is trained with a relatively high batch size. Combined with ensemble methods, this produces competitive results in four counterfactual inference benchmarks: IHDP, NEWS, JOBS, and TWINS. Our results indicate that relatively simple methods might be good enough for counterfactual prediction, with quality constraints coming from hyperparameter tuning. Our analysis indicates that the reason behind the observed phenomenon might be “grokking”, a recently developed theory.
Item Type: | Conference or Workshop Item (Paper) |
---|---|
Divisions: | Faculty of Science and Health Faculty of Science and Health > Computer Science and Electronic Engineering, School of |
SWORD Depositor: | Unnamed user with email elements@essex.ac.uk |
Depositing User: | Unnamed user with email elements@essex.ac.uk |
Date Deposited: | 19 May 2023 11:10 |
Last Modified: | 06 Nov 2024 06:23 |
URI: | http://repository.essex.ac.uk/id/eprint/34514 |
Available files
Filename: Counterfactual_predictions_using_simple_neural_networks.pdf