Statistical Mechanics of Learning in a Large Committee Machine
Schwarze, Holm, Hertz, John A.
–Neural Information Processing Systems
We use statistical mechanics to study generalization in large committee machines. For an architecture with nonoverlapping receptive fields a replica calculation yields the generalization error in the limit of a large number of hidden units.
Neural Information Processing Systems
Dec-31-1993
- Country:
- Asia > Singapore (0.04)
- Europe > Denmark
- Capital Region > Copenhagen (0.05)
- North America > United States
- Maryland > Montgomery County
- Bethesda (0.04)
- New York (0.04)
- Maryland > Montgomery County
- Technology: