Optimizing over trained GNNs via symmetry breaking
–Neural Information Processing Systems
Although GNNs are powerful tools for these "forward" prediction tasks, few works discuss the "backward" (or inverse) problem defined on trained GNNs.
Neural Information Processing Systems
Feb-15-2026, 19:19:31 GMT
- Country:
- Europe
- Germany (0.04)
- United Kingdom > England
- Greater London > London (0.04)
- Asia
- Middle East > Jordan (0.04)
- Myanmar > Tanintharyi Region
- Dawei (0.04)
- Europe
- Genre:
- Research Report (0.46)
- Technology: