Category: presentationvideo
Giovanni Ansaloni: how do you squeeze AI into power-constrained edge devices?
Dr. Giovanni Ansaloni presented a deep dive into methodologies for algorithms optimization at ACACES24 – Signal compression with inputs sparsity and self similarity– Lightweight feature extraction and classification– EdgeML model optimization: few-shot learning personalization distillation federated learning
Digipredict: what is a digital twin?
Flavio Ponzina presents paper on edge AI acceleration at Embedded Systems Week
Dr Flavio Ponzina, who recently defended his PhD thesis on Hardware-Software co-design Methodologies for Edge AI Optimization, presented a paper entitled “Overflow-free compute memories for edge AI acceleration” at Embedded Systems Week, which took place in Hamburg. The video of his presentation is below:
Talk on monitoring and managing data centers with IoT
Prof. David Atienza gave a talk at the Intelligent Maintenance Conference which was held at EPFL in September. Prof. David Atienza examined the inefficient monitoring and management of data centers, and how they can impact both energy costs for the sustainability of our planet and the reliability of data center services. In this talk he (…)
epiPhone – discretely monitoring brain activity
We are proud to present epiPhone, a discreet headset which can monitor brain activity and transmit the data in real time. Today, epilepsy is one of the most common chronic diseases affecting more than 65 million people worldwide. However, no reliable wearable device currently exists for real-time epileptic seizure detection. One of the main challenges (…)
The nervous system as an IoT: David Atienza on brain-inspired healthcare wearables
Android lab course presentation @GamesOnCampus
Giovanni Ansaloni presents Games on Campus: The Lab on Apps Development for Tablets, Smartphones and Smartwatches consists of eight weeks of lab sessions covering everything from the Android operating system to app development, including connections to GPS systems and the cloud. This is followed by a six week project in one of a variety of (…)
A flexible in-memory computing architecture for heterogeneously quantized CNNs
Full Presentation: Self-aware anomaly detection for epilepsy monitoring
Farnaz Forooghifar presents our research on self-aware anomaly detection for epilepsy monitoring on low-power, wearable electrocardiographic devices.