Browsing by Issue Date, starting with "2020-12-21"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Continual learning for object and scene classificationPublication . Turner, Daniel; Rodrigues, J. M .F.; Cardoso, Pedro J. S.Since their existence, computers have been a great asset to mankind, primarily because of their ability to perform specific tasks at speeds humans could never compete with. However, there are many tasks that humans consider easy which are quite difficult for computers to perform. For instance, a human can be shown a picture of an automobile and a bicycle and then be able to easily discriminate between future automobiles and bicycles. For a computer to perform such a task using current algorithms, typically, it must first be shown a large number of images of the two classes, with varying features and positions, and then spend a great deal of time learning to extract and identify features so that it can successfully distinguish between the two. Nevertheless, it is still able to perform the task (eventually) and, after the computational training is complete, would be able to classify images of automobiles and bicycles faster, and sometimes better, than the human. Nonetheless, the real out-performance displayed by the human is when another class is added to the mix, e.g., “aeroplane”. The human can immediately add aeroplanes to its set of known objects, whereas a computer would typically have to go almost back to the start and re-learn all the classes from scratch. The reason the network requires to be retrained is because of a phenomenon named Catastrophic Forgetting, where the changes made to the system during the acquisition of new knowledge bring about the loss of previous knowledge. In this dissertation, we explore Continual Learning, where we propose a way to deal with Catastrophic Forgetting by making a framework capable of learning new information without having to start from scratch and even “improving” its knowledge on what it already knows. With the above in mind, we implemented a Modular Dynamic Neural Network (MDNN) framework, which is primarily made up of modular sub-networks and progressively grows and re-arranges itself as it learns continuously. The network is structured in such a way that its internal components function independently from one another so that when new information is learned, only specific sub-networks are altered in a way that most of the old information is not forgotten. The network is divided into two main blocks, the feature extraction component which is based on a ResNet50 and the modular dynamic classification sub-networks. We have, so far, achieved results below those of the state of the art using ImageNet and CIFAR10, nevertheless, we demonstrate that the framework can meet its initial purpose, which is learning new information without having to start from scratch.
- Understanding new microbial communication systems to combat antimicrobial resistancePublication . Rossetto, Veronica; Galvão, Helena M.; Reen, JerryBacterial biofilms provide an advantageous spatial structure for colonization and cell maintenance as a community. This multicellular behaviour is regulated by a bacterial quorum-dependent mechanism, called the quorum sensing (QS) system that regulates other diverse social behaviours such as toxin production and virulence factors. This mechanism is regulated by signal molecules that regulate intra-specific, inter-specific and inter-kingdom interactions. For these reasons, this mechanism is strongly studied, as well as signalling molecules and analogues, such as alkyl-quinolone based compounds, for the disarming of pathogenic bacteria resistant to multi-drugs that plague public health worldwide. The path to a complete understanding of how this occurs, what are the conditions for such biological responses and what machinery and mechanisms exist for the perception and modulation of these interactions is still far from reaching a conclusion. Therefore, the present work seeks to evaluate compounds against behaviours dependent on the quorum sensing mechanism, as well as the effect of these compounds on the growth of harmful pathogens, such as Pseudomonas aeruginosa and Staphylococcus aureus. Providing information to assist in understanding these microbial interactions, as well as the development of new anti-infectious strategies and the fight against antimicrobial resistance. The tested compounds confirmed activities such as anti-biofilm, anti-swarming and anti-pyocyanin production. Of the twenty-three analogous compounds to the alkyl-quinolones screened thirteen presented some type of interference between the three evaluated phenotypes, five with significant antagonistic activities against P. aeruginosa PA14 and three against staphylococcal strains, such as S. aureus, Staphylococcus haemolyticus and Staphylococcus hominis. Thus, it is concluded that small molecules based on alkyl-quinolones are effective bioactive against QS dependent behaviours and can assist in unravelling microbial communication and its impacts on human society.