Identification of new M31 star cluster candidates from PAndAS images using convolutional neural networks

Context.Identification of new star cluster candidates in M31 is fundamental for the study of M31 stellar cluster system. The machine learning method Convolutional Neural Network (CNN) is an efficient algorithm to search for new M31 star cluster candidates from tens of millions of images from wide-field photometric surveys. Aims.We search for new M31 cluster candidates from the high quality $g$ and $i$-band images of 21,245,632 sources that obtained by the Pan-Andromeda Archaeological Survey (PAndAS) by CNN. Methods.We collect confirmed M31 clusters and non cluster objects from the literature as our training sample. Accurate double-channel CNNs have been constructed and been trained using the training samples. We have applied the CNN classification models to the PAndAS $g$ and $i$-band images of over 21 million sources to search new M31 cluster candidates. The CNN predictions are finally checked by five experienced inspectors to obtain high confidence M31 star cluster candidates. Results.After human-inspection, we have identified a catalogue of 117 new M31 cluster candidates. Most of the new candidates are young clusters being located in the M31 disk. Their morphology, colours and magnitudes are similar as those of the confirmed young disk clusters. We have also identified eight globular cluster candidates which are located in the M31 halo and exhibit features similar as those confirmed halo globular clusters. Three of them have projected distances to the M31 centre larger than 100\,kpc.