Interventional Multi-Instance Learning with Deconfounded Instance-Level Prediction
This addresses robustness and interpretability issues in multi-instance learning for applications like medical imaging, though it is incremental as it builds on existing causal inference concepts.
The paper tackles the problem of bag contextual confounding in multi-instance learning, which affects instance-level prediction accuracy, and proposes an interventional framework that reduces false positives and outperforms state-of-the-art methods in pathological image analysis.
When applying multi-instance learning (MIL) to make predictions for bags of instances, the prediction accuracy of an instance often depends on not only the instance itself but also its context in the corresponding bag. From the viewpoint of causal inference, such bag contextual prior works as a confounder and may result in model robustness and interpretability issues. Focusing on this problem, we propose a novel interventional multi-instance learning (IMIL) framework to achieve deconfounded instance-level prediction. Unlike traditional likelihood-based strategies, we design an Expectation-Maximization (EM) algorithm based on causal intervention, providing a robust instance selection in the training phase and suppressing the bias caused by the bag contextual prior. Experiments on pathological image analysis demonstrate that our IMIL method substantially reduces false positives and outperforms state-of-the-art MIL methods.