On associative neural networks for sparse patterns with huge capacities

ArXi:2603.26217v1 Announce Type: cross Generalized Hopfield models with higher-order or exponential interaction terms are known to have substantially larger storage capacities than the classical quadratic model. On the other hand, associative memories for sparse patterns, such as the Willshaw and Amari models, already outperform the classical Hopfield model in the sparse regime. In this paper we combine these two mechanisms. We of sparse associative memory models and study their storage capacities.