Optimizing Information-theoretical Generalization Bounds via Anisotropic Noise in SGLD
–Neural Information Processing Systems
We then apply matrix analysis to derive the form of optimal noise covariance. Presented constraint and results are validated by the empirical observations.
Neural Information Processing Systems
Nov-15-2025, 20:24:10 GMT
- Country:
- Asia > China (0.04)
- Europe > United Kingdom
- England > Oxfordshire > Oxford (0.04)
- North America > United States
- Massachusetts > Norfolk County > Wellesley (0.04)