AWS – Amazon SageMaker Automatic Model Tuning now reuses SageMaker Training instances to reduce start-up overheads by 20x

Amazon SageMaker Automatic Model Tuning now reduces the start-up time of each training job launched to tune your models by 20x on average (from 2.5 minutes to 8 seconds). In scenarios where you have a large number of hyperparameter evaluations, the reuse of training instances can cumulatively save 2 hours for every 50 sequential evaluations.

Read More for the details.