MARVEL Archives

Fairness and underspecification in acoustic scene classification: The case for disaggregated evaluations

Post published:October 5, 2021
Post category:Publications

Underspecification and fairness in machine learning (ML) applications have recently become two prominent issues in the ML community. Acoustic scene classification (ASC) applications have so far remained unaffected by this discussion, but are now becoming increasingly used in real-world systems where fairness and reliability are critical aspects.

MARVEL: Multimodal Extreme Scale Data Analytics for Smart Cities Environments

Post published:October 5, 2021
Post category:Publications

A Smart City based on data acquisition, handling and intelligent analysis requires efficient design and implementation of the respective AI technologies and the underlying infrastructure for seamlessly analyzing the large amounts of data in real-time.

Improving the Accuracy of Early Exits in Multi-Exit Architectures via Curriculum Learning

Post published:September 22, 2021
Post category:Publications

Deploying deep learning services for time-sensitive and resource-constrained settings such as IoT using edge computing systems is a challenging task that requires dynamic adjustment of inference time. Multi-exit architectures allow deep neural networks to terminate their execution early in order to adhere to tight deadlines at the cost of accuracy.

Automatic Analysis of the Emotional Content of Speech in Daylong Child-Centered Recordings from a Neonatal Intensive Care Unit

Post published:September 22, 2021
Post category:Publications

Researchers have recently started to study how the emotional speech heard by young infants can affect their developmental outcomes. As a part of this research, hundreds of hours of daylong recordings from preterm infants’ audio environments were collected from two hospitals in Finland and Estonia in the context of so-called APPLE study.

Benchmarking solutions within MARVEL

Post published:September 19, 2021
Post category:Blog

Benchmarking can be defined as the way to measure and compare performance metrics or business processes with the current state-of-the-art.

DCASE Challenge: Pushing the state-of-the-art in audio content analysis technologies

Post published:September 19, 2021
Post category:Blog

The core idea of the community is to support the development of audio content analysis methods by providing public datasets and the opportunity to continuously compare different approaches with consistent performance measures in uniform setups.

Improvement of quality of life for kids, youth and elderly

Post published:September 15, 2021
Post category:Blog

How can kids, the youth and the elderly benefit from better knowledge about smart cities? How can they benefit from contributing to smart cities services by using smart devices? How does MARVEL help in improving their quality of life?

Ethics Advisory Board in Horizon Europe: Functions and Role

Post published:August 29, 2021
Post category:Blog

There are various approaches to ethical issues. While some of these issues invite simple solutions, other demand complex, multilayered solutions. Regardless of the favored approach, the actions of the Ethics Advisory Board (EAB) and its Ethics Advisors (EA) must be carried out with the best interests of the general public and civil society in perspective.

MARVEL at AI nella sicurezza urbana a Trento

Post published:August 10, 2021
Post category:News

The event was focused on urban security and a targeted audience of people with some AI affinity.

MARVEL at Forum PA 2021

Post published:August 10, 2021
Post category:News

The Forum PA was held online from June 21st to June 25th, 2021, organized by Gruppo Digital360. The event was held virtually and MARVEL was represented by the Municipality of Trento at a virtual booth.