Bi-linear feature learning models, like the gated autoencoder, were proposed as a way to model relationships between frames in a video. By minimizing reconstruction error of one frame, given the previous frame, these models learn “mapping units” that encode the transformations inherent in a sequence, and thereby learn to encode motion. In this work we extend bi-linear models by introducing “hig...

In this paper, a new version of multi-objective differential evolution with dynamically adaptable mutation factor is used for Pareto optimization of a 5-degree of freedom vehicle vibration model excited by non-stationary random road profile. In this way, non-dominated sorting algorithm and crowding distance criterion have been combined to differential evolution with fuzzified mutation in order ...

Today, transportation of hazardous materials (i.e., hazmat) is a very important issue for researchers due to the risk of this transit, which should be considered for development of industries and transportation. Therefore, a model that is useful should consider the risk and damage to humanitarian and environmental issues due to transit of hazmat materials. By considering the related cost, the r...

In this paper, bi-level linear programming problem with multiple objectives at each level in intuitionistic fuzzy environment is studied and solved. In this problem, technological coefficients and resources are characterized by intuitionistic fuzzy numbers having linear membership and non-membership function. Intuitionistic fuzzy efficient solution is computed for the upper level by the propose...

research interests have been focused on the concept of penalizing jobs both for being early and for being tardy because not only of modern competitive industrial challenges of providing a variety of products at a very low cost by smoothing productions but also of its increasing and exciting computer applications. here, sequencing approaches of the mixed- model just-in-time production systems is...

We present a new exact approach for solving bi-objective integer linear programs. The new approach efficiently employs two of the existing exact algorithms in the literature, including the balanced box and the -constraint methods, in two stages. A computationally study shows that (1) the new approach solves less single-objective integer linear programs in comparison to the balanced box method, ...

classical deterministic inventory models consider the demand rate to be either constant or time-dependent but independent from the stock status. however, for a certain type of inventory, the demand rate may be in-fluenced by the stock level. also in many real-life problems, some products such as fruits, vegetables, phar-maceuticals and volatile liquids continuously deteriorate to evaporation, o...

instabilities of a thin viscous film flowing down a uniformly heated plane are investigated in this study. the heating generates a surface tension gradient that induces thermocapillary stresses on the free surface. thus, the film is not only influenced by gravity and mean surface tension but also the thermocapillary force is acting on the free surface. moreover, the heat transfer at the free su...