An Investigation of the Weight Space to Monitor the Training Progress of Neural Networks
This addresses the need for efficient monitoring of model updates in deployment, though it is incremental as it builds on existing weight space analysis.
The paper tackled the problem of monitoring changes in deep neural networks during training by investigating weight space trajectories, finding that models evolve on unique, smooth paths that can be used to track progress and potentially detect domain shifts.
Safe use of Deep Neural Networks (DNNs) requires careful testing. However, deployed models are often trained further to improve in performance. As rigorous testing and evaluation is expensive, triggers are in need to determine the degree of change of a model. In this paper we investigate the weight space of DNN models for structure that can be exploited to that end. Our results show that DNN models evolve on unique, smooth trajectories in weight space which can be used to track DNN training progress. We hypothesize that curvature and smoothness of the trajectories as well as step length along it may contain information on the state of training as well as potential domain shifts. We show that the model trajectories can be separated and the order of checkpoints on the trajectories recovered, which may serve as a first step towards DNN model versioning.