Bisimulations for Neural Network Reduction
This work addresses the need for efficient neural network compression for practitioners, though it appears incremental as it builds on existing bisimulation concepts.
The authors tackled the problem of neural network reduction by introducing a notion of bisimulation for semantic equivalence and an approximate version for semantic closeness, providing a trade-off between reduction scale and semantic deviation.
We present a notion of bisimulation that induces a reduced network which is semantically equivalent to the given neural network. We provide a minimization algorithm to construct the smallest bisimulation equivalent network. Reductions that construct bisimulation equivalent neural networks are limited in the scale of reduction. We present an approximate notion of bisimulation that provides semantic closeness, rather than, semantic equivalence, and quantify semantic deviation between the neural networks that are approximately bisimilar. The latter provides a trade-off between the amount of reduction and deviations in the semantics.