Gradient Checkpointing for Efficient Backpropagation¶
Authors: Chen et al. Year: 2016 ArXiv/Link: https://arxiv.org/abs/1604.06174
Summary¶
Memory-efficient training technique that trades computation for memory by recomputing activations during backpropagation.
Key Concepts¶
- Memory efficiency
- Recomputation
- Trade-off strategies
- Backpropagation
- Training optimization
Impact¶
Enabled training of large models with limited memory
Category¶
Optimization & Efficiency