In most classification tasks there are observations that are ambiguous and
therefore difficult to correctly label. Set-valued classifiers output sets of
plausible labels rather than a single label, thereby giving a more appropriate
and informative treatment to the labeling of ambiguous instances. We introduce
a framework for multiclass set-valued classification, where the classifiers
guarantee user-defined levels of coverage or confidence (the probability that
the true label is contained in the set) while minimizing the ambiguity (the
expected size of the output). We first derive oracle classifiers assuming the
true distribution to be known. We show that the oracle classifiers are obtained
from level sets of the functions that define the conditional probability of
each class. Then we develop estimators with good asymptotic and finite sample
properties. The proposed estimators build on existing single-label classifiers.
The optimal classifier can sometimes output the empty set, but we provide two
solutions to fix this issue that are suitable for various practical needs.

此研究针对分类问题中存在的模糊不确定标签，提出了一种多类别集合值分类方法，在最小化分类的模糊性的前提下，保证用户定义的覆盖率和置信度，利用条件概率函数的级别集导出 oracle 分类器，并提出一种在现有单标记分类器基础上的估计器，可以在不同的实际应用需求下解决因优化而导致的输出空集问题。

具有有界误差水平的最少歧义集值分类器

Least Ambiguous Set-Valued Classifiers with Bounded Error Levels

Modern day wireless networks have tremendously evolved driven by a sharp
increase in user demands, continuously requesting more data and services. This
puts significant strain on infrastructure based macro cellular networks due to
the inefficiency in handling these traffic demands, cost effectively. A viable
solution is the use of unmanned aerial vehicles (UAVs) as intermediate aerial
nodes between the macro and small cell tiers for improving coverage and
boosting capacity. This letter investigates the problem of user demand based
UAV assignment over geographical areas subject to high traffic demands. A
neural based cost function approach is formulated in which UAVs are matched to
a particular geographical area. It is shown that leveraging multiple UAVs not
only provides long range connectivity but also better load balancing and
traffic offload. Simulation study demonstrate that the proposed approach yields
significant improvements in terms of 5th percentile spectral efficiency up to
38\% and reduced delays up to 37.5\% compared to a ground-based network
baseline without UAVs.

本研究探讨了基于用户需求的无人机分配问题，提出一种基于神经网络成本函数的方法，以无人机作为中间节点，提高网络覆盖率、容量和负载均衡，模拟研究显示该方法相比于传统的地面网络，可有效提高无线电谱效率和降低延迟。