TECoSA Seminar – Efficient Inference at the Edge

We aim to bring you a TECoSA Seminar on the first Thursday of each month during term-time. For Autumn 2022, the talks will be on-line or hybrid. All are welcome to attend and we look forward to some lively discussions. Members can accept the invitations, non-members can email vickid @ kth.se to register.

Our November seminar is with Professor Axel Jantsch, Professor of systems on chips at TU Wien.  The session is co-hosted with Digital Futures, and will be given via Zoom (https://kth-se.zoom.us/j/66857695267), with a number of seats available in real life on KTH Campus (Digital Futures Conference Room, Osquars backe 5, SE-114 28 Stockholm).  Please email vickid @ kth.se by 31 October if you plan to join in real life.

ABSTRACT:  The presentation will review the design space, the trade-offs and the available options for implementing Deep Neural Networks in resource constrained embedded devices.  Going into depth, the talk will present:
~ a detailed, layer-wise power and delay profiling of neural network inference, and
~ pruning based network compression techniques.

BIO: Axel Jantsch received the Ph.D. degree from TU Wien, Vienna, Austria, in 1992. From 2002 to 2014, he was a full professor in      electronic systems design at KTH. Since 2014, he has been a Professor of systems on chips at TU Wien, Vienna, Austria.  His current research interests include systems on chips, self-aware cyber-physical systems, and embedded machine learning.  You can read more on his homepage: https://jantsch.se/AxelJantsch/HomePage/research.html

This event is co-sponsored by Digital Futures (www.digitalfutures.kth.se)