Enabling ultra high speed wireless communication, Extreme High Frequency (EHF) or Millimeter Wave (mmWave) bands will play a significant role for the 5G. Apart from speed, 5G will be very useful for handling great amounts of data simultaneously and serving bandwidth hungry applications as well. Ultra high quality and ultra fast video streaming will be one of those applications that will be made possible by 5G. While serving bandwidth hungry applications with ease will be an important development and maximizing throughput is most of the time the main goal in a network, it is also important to make sure that no other application starves. In order to prevent such a situation, fair bandwidth allocation should be considered in wireless communications. We simulated a max-min fair bandwidth allocation scenario in a mmWave radio cluster, where a radio cluster is a set of base stations connected to a main hub over 60 GHz radio links. We ran experiments with different path loss exponent values with increasing number of base stations to examine the effects of topology complexity and radio signal loss on the optimization time and on the overall network throughput while maintaining max-min fair allocation. The results showed that as the topology becomes more complex, the problem takes longer to solve. However, the overall network throughput increases. In addition, our model has achieved a decent quantitative fairness level as shown by Jain's index values, which are always more than 0.5 on a scale of 0 to 1 with respect to the topology complexity and the number users.

Scene models allow robots to reason about what is in the scene, what else should be in it, and what should not be in it. In this paper, we propose a hybrid Boltzmann Machine (BM) for scene modeling where relations between objects are integrated. To be able to do that, we extend BM to include tri-way edges between visible (object) nodes and make the network to share the relations across different objects. We evaluate our method against several baseline models (Deep Boltzmann Machines, and Restricted Boltzmann Machines) on a scene classification dataset, and show that it performs better in several scene reasoning tasks.