Boolean Logic as an Error feedback mechanism
This addresses the problem of ensuring reliable training for Boolean neural networks, which is incremental as it builds on prior work by adding theoretical guarantees.
The paper tackles the challenge of training neural networks with Boolean weights and activations, which lacks convergence guarantees, by providing the first convergence analysis under standard non-convex assumptions.
The notion of Boolean logic backpropagation was introduced to build neural networks with weights and activations being Boolean numbers. Most of computations can be done with Boolean logic instead of real arithmetic, both during training and inference phases. But the underlying discrete optimization problem is NP-hard, and the Boolean logic has no guarantee. In this work we propose the first convergence analysis, under standard non-convex assumptions.