When modeling high-dimensional richly structured data, it is often the case that the distribution defined by the Deep Boltzmann Machine (DBM) has a rough energy landscape with man...
This paper presents a new supervisory control scheme, which is based on a control-relevant switching logic. Unlike most of the existing switching methods considering only estimato...
In Bayesian machine learning, conjugate priors are popular, mostly due to mathematical convenience. In this paper, we show that there are deeper reasons for choosing a conjugate pr...
Neural-network dynamics frequently organize in assemblies of synchronized neurons that are thought to encode and store sensory information. We have investigated the mechanisms lea...
In this paper, the finite volume method is developed to analyze coupled dynamic problems of nonlinear thermoelasticity. The major focus is given to the description of martensitic ...