Personal tools
You are here: Home Events Nic Lane (Bell Labs)

Nic Lane (Bell Labs)

— filed under:

Squeezing Deep Learning onto Wearables and Phones

  • Colloquium Series
When Nov 27, 2015
from 02:30 PM to 03:30 PM
Where 4.31/4.33
Add event to calendar vCal

Breakthroughs from the field of deep learning are transforming how sensor data (such as images, audio, and even accelerometers and GPS) are interpreted, and the high-level information needed by mobile apps is extracted. The state-of-the-art in computational models for inferring faces, objects, activities, context are increasingly based on the principles and algorithms of deep learning. It is critical that the gains in inference accuracy and robustness that these models afford us become routinely embedded in the emerging sensor-based mobile apps used by consumers. Unfortunately, this is not happening today -- even though mobile apps present some of the most challenging examples of noisy and complex sensor data we face -- in far too many cases, smartphones and wearables use machine learning methods that have been superseded by deep learning years ago.

In this talk, I will describe our recent work in developing general-purpose support for deep learning-based inference on resource-constrained mobile devices. Our goal is to radically lower the mobile resources (such as energy, memory and computation) consumed by these modeling techniques at inference time, removing the key bottleneck preventing the widespread use of these algorithms. The foundation of this research is in the rethinking of how inference algorithms operate under mobile conditions along with increasing the utilization of the complete range of computational units (e.g., DSPs, GPUs, CPUs) now present in devices like watches, glasses and phones. Ultimately in this work, we aim to completely change how mobile sensor data is processed – and in turn, what mobile apps are capable of – in the next generation of personal sensing devices.



Nic Lane is a Principal Scientist at Bell Labs where he is a member of the Internet of Things research group. Before joining Bell Labs, he spent four years as a Lead Researcher at Microsoft Research based in Beijing. Nic received his Ph.D. from Dartmouth College (2011), his dissertation pioneered community-guided techniques for learning models of human behavior. These algorithms enable mobile sensing systems to better cope with diverse user populations and conditions routinely encountered in the real-world. More broadly, Nic's research interests revolve around the systems and modeling challenges that arise when computers collect and reason about people-centric sensor data. At heart, he is an experimentalist who likes to build prototype sensing systems based on well-founded computational models. Results of his research have been publishedin top-tier conferences that focus on ubiquitous computing and mobile sensing research (e.g., AAAI, UbiComp, MobiCom, MobiSys, SenSys).This work has been recognized by the community with best paper awards (UbiComp ’15 and ’12, MobiCASE’12) and best paper nominations(UbiComp ’11 and ‘14). His recent professional service and activities include: serving onthe PC for leading venues in his field(e.g., UbiComp, MobiSys, SenSys, WWW, CIKM), in addition to acting as PC-chair of MobiQuitous 2015 and MobiCASE 2014.

Document Actions