Description
Dynamic Sparse Training method based on using the gradient norm to select which weights to remove/add in each mask update.
Papers Using This Method
Learning Delays Through Gradients and Structure: Emergence of Spatiotemporal Patterns in Spiking Neural Networks2024-07-07RIGL: A Unified Reciprocal Approach for Tracing the Independent and Group Learning Processes2024-06-18Dynamic Sparse Training with Structured Sparsity2023-05-03[Re] Rigging the Lottery: Making All Tickets Winners2021-03-29Rigging the Lottery: Making All Tickets Winners2019-11-25