ReachLipBnB: A branch-and-bound method for reachability analysis of neural autonomous systems using Lipschitz bounds.
Taha EntesariSina SharifiMahyar FazlyabPublished in: CoRR (2022)
Keyphrases
- autonomous systems
- branch and bound method
- reachability analysis
- lower bound
- branch and bound
- upper bound
- branch and bound algorithm
- model checking
- markov decision processes
- lower and upper bounds
- timed automata
- mixed integer programming
- feasible solution
- reduce the search space
- optimal configuration
- incremental algorithms
- dynamic environments
- neural network
- autonomous navigation
- np hard
- state space
- optimal solution
- objective function
- lagrangian relaxation
- tabu search
- special case
- reinforcement learning
- mixed integer
- linear program