Understanding Soft Error Sensitivity of Deep Learning Models and Frameworks through Checkpoint Alteration | Publicación