{"id":31,"date":"2021-05-04T08:13:31","date_gmt":"2021-05-04T06:13:31","guid":{"rendered":"https:\/\/portale.unime.it\/modellopersona\/?page_id=31"},"modified":"2025-04-02T10:07:50","modified_gmt":"2025-04-02T08:07:50","slug":"call","status":"publish","type":"page","link":"https:\/\/portale2.unime.it\/fairground2025\/call\/","title":{"rendered":"CALL FOR PAPERS"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t



CALL FOR PAPERS


<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

CALL FOR PAPERS<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
\n\t\t\t<\/path><\/svg>\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

ABSTRACT<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Borrowed from edge computing, on-device training is a novel paradigm that offers several advantages in terms of latency and security and gives rise to new types of architectures that form the core of modern smart environments consisting of intelligent devices capable of self-learning from their environment. Most of the challenges encountered when training on devices are related to the limited resources of these systems. To overcome this problem, various techniques have been proposed in the literature, such as: compression\/quantization of weights, federated learning and transfer learning. On the other hand, although these approaches allow for a less costly training procedure, they come with requirements that cannot be met in some cases. Given these limitations, neuromorphic solutions such as Spiking Neural Networks (SNNs) have emerged as a promising alternative. They represent a new paradigm in the field of neural computation and offer a unique approach to modeling and understanding the intricacies of information processing in the brain. Neuromorphic perception and processing mimics the asynchronous and sparse nature of information processing in animal brains. This is indeed a potential technological breakthrough with promising prospects for autonomous systems such as robots and implantable systems, where energy and size play a crucial role. Neuromorphic computing, which aims to bridge the gap between neuroscience and artificial intelligence, uses the principles underlying neurobiological systems to process data more efficiently than conventional architectures. In doing so, it overcomes the challenges typically encountered when working with artificial neural networks (ANNs) in terms of energy efficiency, event-driven computation and adaptation to environmental changes, making it suitable for a variety of application domains including robotics. This workshop aims to explore the extensive use of neuromorphic computing systems, also in combination with artificial networks, for the implementation of novel bio-inspired architectures and applications by exploiting innovative approaches and optimizing computational resources.<\/p>

The topics of interest include, but are not limited to:<\/p>