Invited Lecture: Bringing Fast Deep Learning to Mobiles through Approximate Computing

Loading Events

Our reliance on smartphones demands continual advancement of mobile computing. Yet, our computing appetites grow much faster than the current hardware technology advances, producing a critical strain on the mobile’s limited resources. To address this issue, we propose Approximate Mobile Computing (AMC) and take a radical stance that computation does not need to be 100% precise. We first examine situations, such as mobile video playback and mobile deep learning for human activity recognition, where the properties of the input and the limitations of human perception open space for AMC. We then develop methods that bring AMC to consumer devices, including an Android compiler framework that enables dynamic tuning of the level of approximation according to the context of usage. Finally, we look into the future of AMC for efficient mobile sensing and model training as well.
Speaker(s): Veljko Pejovic, PhD,
Room: Meeting Room, Bldg: FEEIT, Rugjer Boshkovikj 18, Postal box 574, Skopje, Macedonia, Macedonia, 1000

Share This Story, Choose Your Platform!

Go to Top