Downlink
scheduling in the LTE system is an open problem for which several heuristic
solutions exist. Recently, there has been an increase in interest in ap-plying machine learning to networking problems,
including downlink scheduling. Improvements in Physical Layer capabilities have
generated new resource-intensive use cases and continuously modifying existing
heuristic solutions could result in the development of systems too complex to
maintain over time. We propose a LSTM/Pointer Network-based downlink scheduler which aims to improve upon the current models
which utilize feed forward neural networks. Our scheduler flex-ibly handles changing numbers of UEs via the use of a
recurrent neural network. We formulate the downlink
scheduling problem as a Markov Decision Process that integrates the
channel quality indicator and the buffer size of each UE as the
observation and solve it using a Deep Reinforcement Learning algorithm. Our experiments demonstrate that our approach results in a
scheduler which gener-alised across changing number
of UEs and resource blocks and performed within the range of traditional
schedulers.