PVF: A novel metric for understanding AI systems’ vulnerability against SDCs in model parameters

We’re introducing parameter vulnerability factor (PVF), a novel metric for understanding and measuring AI systems’ vulnerability against silent data corruptions (SDCs) in model parameters. PVF can be tailored to different AI models and tasks, adapted to different hardware faults, and even extended to the training phase of AI models. We’re sharing results of our own [...]


Read More...


The post PVF: A novel metric for understanding AI systems’ vulnerability against SDCs in model parameters appeared first on Engineering at Meta.


http://dlvr.it/T8VMWQ

Komentar

Postingan populer dari blog ini

How Meta enforces purpose limitation via Privacy Aware Infrastructure at scale

Fully Sharded Data Parallel: faster AI training with fewer GPUs

Risk-driven backbone management during COVID-19 and beyond