ReSiN resets neurons that have both low forward output and low backward gradient to preserve RL plasticity in streaming, but the proof of the central characterization is incorrect.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Silent Neuron Theory and Plasticity Preservation for Deep Reinforcement Learning in Adaptive Video Streaming
ReSiN resets neurons that have both low forward output and low backward gradient to preserve RL plasticity in streaming, but the proof of the central characterization is incorrect.