A neural Shortest Path Network approximates Gibbs-sampled routes to make joint facility-location and path optimization scalable, with roughly 6% path-cost gap and large speedups.
Further, repeating these operations for each1≤k≤Mresults in a total ofO(M 4)operations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Parametrized Multi-Agent Routing via Deep Attention Models
A neural Shortest Path Network approximates Gibbs-sampled routes to make joint facility-location and path optimization scalable, with roughly 6% path-cost gap and large speedups.