Picuslab-DIETI

22/12/2023

12/08/2023

Picuslab-DIETI is happy to share its last publication entitled "Playing With a Multi Armed Bandit to Optimize Resource Allocation in Satellite-Enabled 5G Networks" - whose authors are Antonio Galli, Vincenzo Moscato, Simon Pietro Romano and Giancarlo Sperlì - has been accepted on IEEE Transactions on Network and Service Management.

In this paper, we address issues associated with the effective management of handover events in satellite-enabled 5G network infrastructures.
Namely, we devise a Combinatorial Multi-agent Multi-Armed Bandit for dynamically allocating 5G gNB available resources over time under uncertainty conditions in the presence of a constellation of LEO satellites, based on several parameters collected and dispatched by an ad hoc orchestration platform.

https://lnkd.in/dhFtzz56

This link will take you to a page that’s not on LinkedIn

22/07/2023

Picuslab-DIETIPicuslab is happy to share its last publication entitled " Few-shot Named Entity Recognition: definition, taxonomy and research directions" - whose authors are Vincenzo Moscato, Marco Postiglione and Giancarlo Sperlì - has been accepted on ACM Transactions on Intelligent Systems and Technology.

In this survey, starting from a clear definition and description of the few-shot NER (FS-NER) problem, we take stock of the current state-of-the-art and propose a taxonomy which divides algorithms in two macro-categories according to the underlying mechanisms: model-centric and data-centric. For each category, we line-up works as a story to show how the field is moving towards new research directions. Eventually, techniques, limitations and key aspects are deeply analyzed to facilitate future studies.

https://lnkd.in/dpKSr_7b

Recent years have seen an exponential growth (+98% in 2022 w.r.t. the previous year) of the number of research papers in the few-shot learning field, which aims to train machine learning models with extremely limited available data. The research interest ...

14/06/2023

Picuslab-DIETI is happy to share its last publication entitled "A community detection approach based on network representation learning for repository mining" - whose authors are Marco De Luca, Anna Rita Fasolino, Antonino Ferraro, Vincenzo Moscato, Giancarlo Sperlì, Porfirio Tramontana - has been accepted on Elsevier Expert Systems with Applications Journal.

In this paper, we propose a novel heterogeneous graph-based model for capturing and handling all the complex and strongly-correlated information of a software Developer Social Network (DSN) to support several analytic tasks. In particular, we challenge the problem of automatically discovering communities of software developers sharing interests for similar projects by relying on Social Network Analysis (SNA) findings. To overcome the huge graph-size issue, we leverage different graph embedding techniques.

https://lnkd.in/dq8V5igS

28/05/2023

Picuslab-DIETI is pleased to announce that Professor Sansone has spoken at the entitled Seminar "False Truths" in the Forum PA 2023 event scheduled for May 16-18, 2023.

https://www.youtube.com/watch?v=z9Bw4AIaHlw

16 MAGGIO 2023 - FORUM PA

18/05/2023

We at Picuslab-DIETI are excited to announce the publication of our latest research paper titled "Transformers in the Real World: A Survey on NLP Applications" in the MDPI Information. This paper presents an in-depth survey of the transformative influence of Transformers in the Natural Language Processing (NLP) domain since its inception in 2017. Our investigation highlights the open-access and real-world applications of Transformers, specifically where text is the primary modality.

The paper can be accessed here: https://lnkd.in/d3kTQP2j

The field of Natural Language Processing (NLP) has undergone a significant transformation with the introduction of Transformers. From the first introduction of this technology in 2017, the use of transformers has become widespread and has had a profound impact on the field of NLP. In this survey, we...

20/04/2023

Picuslab-DIETI is happy to share its last publication entitled "Multi-task learning for few-shot biomedical relation extraction" - whose authors are Vincenzo Moscato, Giuseppe Napolano, Marco Postiglione and Giancarlo Sperlì - has been accepted on Artificial Intelligence Review.

In this paper, we propose a framework, built upon a state-of-the-art multi-task method (i.e. MT-DNN), that leverages different publicly available biomedical datasets to enhance relation extraction performance. Our model employs a transformer-based architecture with shared encoding layers across multiple tasks, and task-specific classification layers to generate task-specific representations.

https://lnkd.in/drJqzGPM

Artificial intelligence (AI) has advanced rapidly, but it has limited impact on biomedical text understanding due to a lack of annotated datasets (a.k.a. few-shot learning). Multi-task learning, which uses data from multiple datasets and tasks with related syntax and semantics, has potential to addr...

12/03/2023

Picuslab-DIETI is happy to share its last publication entitled "ReUse:
REgressive Unet for Carbon Storage and Above-Ground Biomass
Estimation",- whose authors are Antonio Elia Pascarella, Giovanni
Giacco, Mattia Rigiroli, Stefano Marrone and Carlo Sansone - has been accepted on Journal of Imaging.

In this work, we build a deep-learning approach that estimates forest carbon absorption using satellite images and public data. It outperforms existing methods, making it a powerful tool for the early detection of carbon absorption changes in urban and rural areas.

https://lnkd.in/dk9YrwvT

The United Nations Framework Convention on Climate Change (UNFCCC) has recently established the Reducing Emissions from Deforestation and forest Degradation (REDD+) program, which requires countries to report their carbon emissions and sink estimates through national greenhouse gas inventories (NGHG...

11/02/2023

Picuslab-DIETI is happy to share its last publication entitled "TaughtNet: Learning Multi-Task Biomedical Named Entity Recognition From Single-Task Teachers" - whose authors are Vincenzo Moscato, Marco Postiglione, Carlo Sansone and Giancarlo Sperlì - has been accepted on IEEE Journal of Biomedical and Health Informatics.

In this work, we propose TaughtNet , a knowledge distillation-based framework allowing us to fine-tune a single multi-task student model by leveraging both the ground truth and the knowledge of single-task teachers .

In Biomedical Named Entity Recognition (BioNER), the use of current cutting-edge deep learning-based methods, such as deep bidirectional transformers (e.g. BERT, GPT-3), can be substantially hampered by the absence of publicly accessible annotated datasets. When the BioNER system is required to anno...

07/02/2023

Picuslab-DIETI is happy to share its last publication entitled "Covid-19 sentiment analysis based on Tweets" - whose authors are Valerio La Gatta, Vincenzo Moscato, Marco Postiglione and Giancarlo Sperlì - has been accepted on IEEE Intelligent System Journal.

In this work, we investigate how individuals in Italy perceived the COVID-19 outbreak and its implications in real-life. Our analysis shows that while the overall sentiment is negative, Italians have shown upbeat responses to the pandemic, especially in regards to the vaccination campaign. The emotion analysis reveals that while fear progressively decreased after the first wave of the pandemic, the overall anger has remained constant but gradually turned into various narratives.

https://lnkd.in/dNUG4ghc

-19

This link will take you to a page that’s not on LinkedIn

07/02/2023

Picuslab-DIETI is happy to share its last publication entitled "A deep attention based approach for predictive maintenance applications in IoT scenarios" - whose authors are Roberto De Luca, Antonino Ferraro, Antonio Galli, Mosè Gallo, Vincenzo Moscato, Giancarlo Sperlì - has been accepted on Journal of Manufacturing Technology Management.

In this paper, a deep learning-based approach has been designed for the predictive maintenance task. Its main novelty is to leverage a multi-head attention (MHA) mechanism for improving RUL estimation and reducing memory requirements.

https://lnkd.in/dwCErZvS

.0

A deep attention based approach for predictive maintenance applications in IoT scenarios - Author: Roberto De Luca, Antonino Ferraro, Antonio Galli, Mosè Gallo, Vincenzo Moscato, Giancarlo Sperlì

25/01/2023

Picuslab-DIETI is happy to share its last publication entitled "An action–reaction influence model relying on OSN user-generated content" - whose authors are Aniello De Santo, Antonino Ferraro, Vincenzo Moscato and Giancarlo Sperlì - has been accepted on Knowledge and Information Systems Journal.

In this paper, we enrich the influence maximization task with a psychological dimension and define a model that ties influence diffusion to recurrent users’ behavior from OSN logs, considering relationships between users mediated by user-generated content.

Due to the sustained popularization of Online Social Networks (OSNs), it has become of interest for a variety of domains of applications to correctly characterize how the behavior of an individual user can be influenced by the actions of other users in a network. Additionally, the richness of availa...