On the Probability of Necessity and Sufficiency of Explaining Graph Neural Networks: A Lower Bound Optimization Approach.
Ruichu CaiYuxuan ZhuXuexin ChenYuan FangMin WuJie QiaoZhifeng HaoPublished in: CoRR (2022)
Keyphrases
- lower bound
- neural network
- upper bound
- min sum
- pattern recognition
- highly non linear
- directed graph
- graph model
- branch and bound algorithm
- objective function
- graph representation
- graph theory
- branch and bound
- optimization algorithm
- random walk
- optimal solution
- global optimization
- fuzzy logic
- lower and upper bounds
- connected components
- optimization problems
- recurrent neural networks
- worst case
- probability distribution
- arbitrarily close
- neural nets
- graph mining
- randomized algorithm
- artificial neural networks