ML on the Edge: Hardware and Models for Machine Learning on Constrained Platforms
Deep neural networks are a key technology at the core of advanced audio and video applications. As these applications begin to migrate from large servers executing in the cloud to mobile and embedded platforms, they place significant demands on the underlying hardware platform. This talk will review the key properties of these models and how these properties are leveraged to deliver efficient inference on energy, compute, and space constrained platforms.
Matthew Mattina is Senior Director of Machine Learning & AI Research at Arm, where he leads a team of researchers developing advanced hardware, software, and algorithms for machine learning. Prior to joining Arm in 2015, Matt was Chief Technology Officer at Tilera, responsible for overall company technology, processor architecture and strategy. Prior to Tilera, Matt was a CPU architect at Intel and invented and designed the Intel Ring Uncore Architecture. Matt has been granted over 30 patents relating to CPU design, multicore processors, on-chip interconnects, and cache coherence protocols. Matt holds a BS in Computer and Systems Engineering from Rensselaer Polytechnic Institute and an MS in Electrical Engineering from Princeton University.