The ubiquitous cellular network is a strong candidate for providing UAVs' wireless connectivity. Due to the maneuverability advantage and higher altitude, UAVs could have line-of-sight (LoS) connectivity with more base station (BS) candidates than terrestrial users. However, the LoS connectivity could also enhance the propagation of up-link interference caused by UAVs over co-existing terrestrial users. In addition, UAVs would perform more handovers than terrestrial users when moving due to the extensive overlap in the coverage areas of many BS candidates. The solution is to bypass the overlapping coverage areas by designing the UAVs' trajectory and to reduce interference by optimizing radio resource allocation through handover management. This paper studies the joint optimization of a UAV's trajectory design and handover management to minimize the weighted sum of three key performance indicators (KPIs): delay, up-link interference, and handover numbers. A dueling double deep Q-network (D3QN) based reinforcement learning algorithm is proposed to solve the optimization problem. Results show that the proposed approach can reduce the handover numbers by 90% and the interference by 18% at the cost of a small increment in transmission delay when compared with the benchmark scheme, which controls the UAV to move along the shortest path and perform handover based on received signal strength. Finally, we verify the advantage of introducing trajectory design, which can reduce the interference by 29% and eliminate the handover numbers by 33% when compared to the D3QN-based policy without trajectory design.

Geo-distributed machine learning (ML) often uses large geo-dispersed data collections produced over time to train global models, without consolidating the data to a central site. In the parameter server architecture, “workers” and “parameter servers” for a geo-distributed ML job should be strategically deployed and adjusted on the fly, to allow easy access to the datasets and fast exchange of the model parameters at anytime. Despite many cloud platforms now provide volume discounts to encourage the usage of their ML resources, different geo-distributed ML jobs that run in the clouds often rent cloud resources separately and respectively, thus rarely enjoying the benefit of discounts. We study an ML broker service that aggregates geo-distributed ML jobs into cloud data centers for volume discounts via dynamic online placement and scaling of workers and parameter servers in individual jobs for long-term cost minimization. To decide the number and the placement of workers and parameter servers, we propose an efficient online algorithm which first decomposes the online problem into a series of one-shot optimization problems solvable at each individual time slot by the technique of regularization, and afterwards round the fractional decisions to the integer ones via a carefully-designed dependent rounding method. We prove a parameterized-constant competitive ratio for our online algorithm as the theoretical performance analysis, and also conduct extensive simulation studies to exhibit its close-to-offline-optimum practical performance in realistic settings.