In this paper we investigate whether Deep Convolutional Neural Networks (DCNNs), which have obtained state of the art results on the ImageNet challenge, are able to perform equally well on three different art classification problems. In particular, we assess whether it is beneficial to fine tune the networks instead of just using them as off the shelf feature extractors for a separately trained softmax classifier. Our experiments show how the first approach yields significantly better results and allows the DCNNs to develop new selective attention mechanisms over the images, which provide powerful insights about which pixel regions allow the networks successfully tackle the proposed classification challenges. Furthermore, we also show how DCNNs, which have been fine tuned on a large artistic collection, outperform the same architectures which are pre-trained on the ImageNet dataset only, when it comes to the classification of heritage objects from a different dataset.

Capabilities of smartphones can be utilised to monitor a range of aspects of users' behaviour. This has potential to affect a number of areas where users' behaviour is considered relevant information. Most notably, healthcare in general and mental health in particular are excellent candidates to utilise capabilities of smartphones, since mental disorders typically have a strong behaviour component. This is especially true for bipolar disorder, where mobility and activity of the patients is considered an indicator of a bipolar episode (depressive or manic). In this work we report on results of using capabilities of smartphones to monitor mobility of the patients, monitored over the period of 12 weeks. Through the continuous discovery of Wi-Fi access points we have inferred significant places (where the patient spent majority of the time) for each patient and investigate correlation of these places with patients' self-reported state. The results show that for majority of patients there exists negative correlation between time spent in clinic and their self-assessment score, while there is a positive correlation between self-assessment scores and time spent outside the home or clinic.

We study the generalization properties of pruned neural networks that are the winners of the lottery ticket hypothesis on datasets of natural images. We analyse their potential under conditions in which training data is scarce and comes from a non-natural domain. Specifically, we investigate whether pruned models that are found on the popular CIFAR-10/100 and Fashion-MNIST datasets, generalize to seven different datasets that come from the fields of digital pathology and digital heritage. Our results show that there are significant benefits in transferring and training sparse architectures over larger parametrized models, since in all of our experiments pruned networks, winners of the lottery ticket hypothesis, significantly outperform their larger unpruned counterparts. These results suggest that winning initializations do contain inductive biases that are generic to some extent, although, as reported by our experiments on the biomedical datasets, their generalization properties can be more limiting than what has been so far observed in the literature.

No abstract

No abstract