Awadelrahman Mohamedelsadig Ali Ahmed

This paper aims to contribute in bench-marking the automatic polyp segmentation problem using generative adversarial networks framework. Perceiving the problem as an image-to-image translation task, conditional generative adversarial networks are utilized to generate masks conditioned by the images as inputs. Both generator and discriminator are convolution neural networks based. The model achieved 0.4382 on Jaccard index and 0.611 as F2 score

Abstract—Non-intrusive load monitoring (NILM) objective is to disaggregate the total power consumption of a building into individual appliance-level profiles. This gives insights to consumers to efficiently use energy and realizes smart grid efficiency outcomes. While many studies focus on achieving accurate models, few of them address the models generalizability. This paper proposes two approaches based on generative adversarial networks to achieve high-accuracy load disaggregation. Concurrently, the paper addresses the model generalizability in two ways, the first is by transfer learning by parameter sharing and the other is by learning compact common representations between source and target domains. This paper also quantitatively evaluate the worth of these transfer learning approaches based on the similarity between the source and target domains. The models are evaluated on three open-access datasets and outperformed recent machine learning methods.

Generative adversarial networks framework immensely attracted the machine learning community attention in the recent years. GANs have succeeded in generating realistic-looking data from noise and have countless applications. On the other hand, Transfer Learning is a smart way that aims to enhance machine learning models in the case of scarce data. The main intuition behind transfer learning is to train models on a plenty of data that is “not” our targeted one but has some similar attributes to it. Then transfer learning in models that contains single networks, such as CNNs, are implemented by sharing the first layers (from the input side) of the model between source and target domains and initialize other layers then retrain the model on our scarce target data. This allows the model to have a good start when learning on the target data and achieve better results than if we trained it from scratch on the target. Now, GANs framework consists of two networks, a generator and a discriminator, and a legit question here is: which network and which part of that network contains the transferable features and hence parameter? Should we fine-tune or freeze the shared parameters? I am trying to answer this specific and direct question by showing imperial results of a series of experiments that I designed based on conditional GANs framework.