This letter studies learning of the binary neural network and its relation to the logical synthesis. The network has the signum activation function and can approximate a desired Boolean function if parameters are selected suitably. In a parameter subspace the network is equivalent to the disjoint canonical form of the Boolean functions. Outside of the subspace, the network can have simpler structure than the canonical form where the simplicity is measured by the number of hidden neurons. In order to realize effective parameter setting, we present a learning algorithm based on the genetic algorithm. The algorithm uses the teacher signals as the initial kernel and tolerates a level of learning error. Performing basic numerical experiments, the algorithm efficiency is confirmed.

In the visual secret sharing scheme proposed by Naor and Shamir, a secret image is encoded into shares, of which size is larger than that of the secret image and the shares are decoded by stacking them without performing any cryptographic computation. In this paper we propose a (k,n) visual secret sharing scheme to encode a black-and-white image into the same size shares as the secret image, where the reconstructed image of the proposed scheme is visible as well as that of the conventional scheme.

Promising proposals of a material, deposition process, and storage device have been demonstrated for neuromorphic systems. The material is Ga-Sn-O (GTO), amorphous metal-oxide semiconductor, and does not contain rare metals such as In. The deposition process is a mist chemical-vapor-deposition (CVD) method, atmospheric pressure process. Therefore, the material and fabrication costs can be simultaneously saved, and three-dimensional stacked structures will be possible. The storage device is an analog memristor, a kind of memristors, but has continuous conductance, and analog computing will be possible owing to continuous weights of synapse elements in neural networks. These structures and computing are the same as those in living brains. We have succeeded in attaining an analog memristive characteristic by optimizing the Ga:Sn composition rate, namely, completing an analog memristor. The analog memristor of the GTO thin film by the mist CVD method can be expected to be a key component for neuromorphic systems.