Publications by authors named "Lorenzo Bonicelli"

Article Synopsis
  • The study investigates the potential distribution of a mosquito species known for spreading vector-borne diseases in central Italy using a combination of entomological data and satellite imagery from Sentinel-2.
  • Three predictive models were utilized: a baseline deep convolutional neural network (DCNN) for environmental conditions at the time of collection, a multitemporal model analyzing conditions over the past two months, and a MultiAdjacency Graph Attention Network (MAGAT) model that considers spatial and climatic relationships.
  • Results showed that the baseline model had a high F1 score of 75.8%, which improved to 81.4% with the multitemporal model and reached 80.9% with the MAGAT model, confirming the widespread presence of the mosquito
View Article and Find Full Text PDF

The staple of human intelligence is the capability of acquiring knowledge in a continuous fashion. In stark contrast, Deep Networks forget catastrophically and, for this reason, the sub-field of Class-Incremental Continual Learning fosters methods that learn a sequence of tasks incrementally, blending sequentially-gained knowledge into a comprehensive prediction. This work aims at assessing and overcoming the pitfalls of our previous proposal Dark Experience Replay (DER), a simple and effective approach that combines rehearsal and Knowledge Distillation.

View Article and Find Full Text PDF

The slaughterhouse can act as a valid checkpoint to estimate the prevalence and the economic impact of diseases in farm animals. At present, scoring lesions is a challenging and time-consuming activity, which is carried out by veterinarians serving the slaughter chain. Over recent years, artificial intelligence(AI) has gained traction in many fields of research, including livestock production.

View Article and Find Full Text PDF