Timezone: »

Locally Adaptive Label Smoothing Improves Predictive Churn
Dara Bahri · Heinrich Jiang

Thu Jul 22 08:50 PM -- 08:55 PM (PDT) @ None

Training modern neural networks is an inherently noisy process that can lead to high \emph{prediction churn}-- disagreements between re-trainings of the same model due to factors such as randomization in the parameter initialization and mini-batches-- even when the trained models all attain similar accuracies. Such prediction churn can be very undesirable in practice. In this paper, we present several baselines for reducing churn and show that training on soft labels obtained by adaptively smoothing each example's label based on the example's neighboring labels often outperforms the baselines on churn while improving accuracy on a variety of benchmark classification tasks and model architectures.

Author Information

Dara Bahri (Google Research)
Heinrich Jiang (Google Research)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors