Gradient Clipping
A safety training strategy that caps maximum gradient values to prevent numerical instability.
Gradient Clipping limits gradient sizes during backpropagation by scaling them when their norm exceeds a specified threshold. This prevents large weight updates that can destabilize training.
Historical figures and technical concepts for informational purposes only. Not technical, professional, legal, or financial advice. Sources: Official Documentation.