Time-Independent Information-Theoretic Generalization Bounds for SGLD
–Neural Information Processing Systems
We provide novel information-theoretic generalization bounds for stochastic gradient Langevin dynamics (SGLD) under the assumptions of smoothness and dissi-pativity, which are widely used in sampling and non-convex optimization studies.
Neural Information Processing Systems
Feb-8-2026, 11:15:37 GMT
- Country:
- Asia
- Afghanistan > Parwan Province
- Charikar (0.04)
- Japan > Honshū
- Kansai > Osaka Prefecture > Osaka (0.04)
- Afghanistan > Parwan Province
- Europe > United Kingdom
- England > Cambridgeshire > Cambridge (0.04)
- Asia
- Technology: