Loughborough University is a top-ten rated university in England for research intensity (REF2014) and an outstanding 66% of the work of Loughborough’s academic staff who were eligible to be submitted to the REF was judged as ‘world-leading’ or ‘internationally excellent’, compared to a national average figure of 43%. In choosing Loughborough for your research, you’ll work alongside academics who are leaders in their field. You will benefit from comprehensive support and guidance from our Doctoral College, including tailored careers advice, to help you succeed in your research and future career. Deep Learning (DL) systems first need to be trained on existing data sets before inference operations can be performed on unseen examples. Most current realisations of (DL) systems target graphical processing units (GPUs), but there has been growing interest in using FPGAs as a hardware platform, since their flexible architecture offers the opportunity of being reconfigured according to the application at hand. A number of FPGA inference solutions have been demonstrated by vendors and researchers, but FPGA training is only just starting to be explored. This research work will consider alternative approaches for the training of FPGA-based DL systems and compare the performance of these approaches in a small number of applications.