Coupled Training with Privileged Information and Unlabeled Data
For practitioners who have extra training-only data that is unreliable, this method provides a way to avoid performance degradation from two-stage approaches.
The paper tackles the problem of leveraging privileged information (available only during training) to improve a deployment model, especially when that information is weak or noisy. They propose a joint training method that learns both models together, which robustly outperforms standard two-stage baselines on synthetic and real-world tasks.
In many prediction problems, we have extra information during training (for example, measurements that are expensive or slow to collect) that will not be available when the model is deployed. A common strategy is to first train a model that uses all training information, then use its predictions on unlabeled examples to train a second model that only uses the inputs available at test time. However, when the extra training-only information is weak or noisy, this Two-Stage approach can mislead the deployment model and even hurt accuracy. We propose a joint training method that learns the two models together, so the deployment model can benefit from the extra information only when it actually helps, instead of inheriting its mistakes. We provide guarantees that describe when joint training improves prediction accuracy and analyze a simple alternating training algorithm for large, high-dimensional models. Experiments on synthetic data and real-world prediction tasks show that our approach avoids these failures and robustly outperforms standard Two-Stage baselines.