2019 European Conference on Ambient Intelligence

The European Conference on Ambient Intelligence is the prime venue for research on Ambient Intelligence, with an international and interdisciplinary character. It brings together researchers and practitioners from the fields of science, engineering, and, design working towards the vision of Ambient Intelligence.

AmI2019 will be held in Rome, Italy at the Sapienza University of Rome on 13-15 November, 2019.

AmI2019 will provide a meeting point for each of these communities, aiming for intensive networking and scientific debate, and shaping visions of the future. This year's event focus topic is “Data-driven Ambient Intelligence” that follows the vision of Calm Technology, where technology is useful but does not demand our full attention or interfere with our usual behavior and activities.

The Proceedings of AmI2019 will be published by Springer in the Lecture Notes in Computer Science (LNCS) series.

AmI 2019 builds on the success of thirteen predecessor conferences, which started in 2003 with the EUSAI-event in Veldhoven, The Netherlands. More information about the AmI series can be found here: http://ami-conferences.org

Invited Speakers

Best Papers Awards

The following papers were selected by the Program Committee based on their scientific merit and also the quality of the presentations given during AmI 2019. The awards were announced during the closing session of the conference.

Workshops

The joint proceedings of the Poster and Workshop Sessions of AmI-2019, are published by CEUR-WS (urn:nbn:de:0074-2492-8) and available ONLINE at the http://ceur-ws.org/Vol-2492/

Tutorial on Embedded Artificial Intelligence

AmI2019 hosts a one-day tutorial organized by STMicroelectronics on Microcontrollers and Artificial Intelligence that will take place during Wednesday, 13th of November, 2019.

The tutorial will focus on how a pre-trained neural network model output from a broad range of the most popular AI frameworks (including Caffe, CNTK, Keras, Lasagne, TensorFlow, and theano), can be maped to an optimized DNN that is adapted to the memory and processing-power capabilities of a target STM32 microcontroller. Techniques on how to check the functionality of the adapted DCNN – which can be 10x smaller than the original, with negligible loss of accuracy.

Check this page for more information on the technical details of the STM toolchain from ST’s AI experts.

The tutorial will include a practical session with a microcontroller board and your laptop. Participants are kindly requested to make sure that the complete tool chain that will be used during the tutorial is properly installed on their laptop.

Participants are encouraged to bring their pre-trained feed forward neural network targeting STM32 resources in Keras or Caffe as a challenge to get it mapped and running.