Reducing Neural Network Parameter Initialization Into an SMT Problem
This addresses the challenge of effective initialization for deep neural networks, though it appears incremental as it builds on prior work with different activations and network depths.
The paper tackles the problem of improving neural network performance by initializing parameters using an SMT solver, achieving better results compared to random initialization.
Training a neural network (NN) depends on multiple factors, including but not limited to the initial weights. In this paper, we focus on initializing deep NN parameters such that it performs better, comparing to random or zero initialization. We do this by reducing the process of initialization into an SMT solver. Previous works consider certain activation functions on small NNs, however the studied NN is a deep network with different activation functions. Our experiments show that the proposed approach for parameter initialization achieves better performance comparing to randomly initialized networks.