Neural Networks

Containing Spillover with Minimal Supervision
Winner of 3rd place in the SPAR Symposium 2024 poster competition.
May 21, 2025
Ariana Azarbal, Matthew A. Clarke, Jorio Cocola, Cailley Factor, Alex Cloud
Project
Containing Spillover with Minimal Supervision
Gradient Routing
Can we control where learning happens in neural networks? Gradient routing addresses this by applying masks to limit the flow of gradients during backpropagation. By supplying different masks for different data points, the user can induce specialized subcomponents within a model. As part of SPAR, I am working on applying this method to AI safety research.
Matthew A. Clarke
Cloud et al., 2025 Cloud et al., 2024
Gradient Routing