Mistake-gated plasticity reduces neural network updates by 50-80% by gating changes on classification errors, improving efficiency for continual learning without added hyperparameters.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Restricting plasticity to large-magnitude updates and to connected paths through the network reduces energy cost of training with only modest extra time.
citing papers explorer
-
Mistake gating leads to energy and memory efficient continual learning
Mistake-gated plasticity reduces neural network updates by 50-80% by gating changes on classification errors, improving efficiency for continual learning without added hyperparameters.
-
Competitive plasticity to reduce the energetic costs of learning
Restricting plasticity to large-magnitude updates and to connected paths through the network reduces energy cost of training with only modest extra time.