Published by the American Physical Society through the American Institute of Physics
Physical Review E, Statistical, nonlinear, and soft matter physics vol:53 issue:1 pages:1276-1279
We discuss appropriate modifications of the Hebbian learning rule for Q-state Potts neural networks with biased patterns, the purpose being to prevent the storage capacity from decreasing drastically with increasing bias. Several prescriptions are compared. As an illustration their retrieval performance is studied numerically for Q = 3.