Program

9:00-9:15

Introductory remarks

9:15-9:45
  • Efficient Methods and Hardware for Deep Neural Networks
  • William Dally
  • Stanford-Nvidia
9:45-10:15
  • Processor Architectures for Deep Learning
  • Amir Khosrowshahi
  • Intel (Nervana)
10:15-11:00

Poster presentations and coffee break

11:00-11:30
  • Deep Learning on Resource-Constrained Devices
  • Ali Farhadi
  • University of Washington and the Allen Institute for Artificial Intelligence
11:30-12:00

Oral Presentations (Session A)

12:00-13:30

Lunch (on your own)

13:30-14:00
  • Joint Design of Algorithms and Hardware for Energy-efficient DNNs
  • Vivienne Sze
  • MIT
14:00-14:30
  • From Training Low Precision Neural Nets to Training Analog Continuous-Time Machines
  • Yoshua Bengio
  • Université de Montréal
14:30-15:30

Poster Presentations and coffee break

15:30-16:00
  • High-Performance Deep Learning
  • Kurt Keutzer
  • Berkeley
16:00-16:30

Oral Presentations (Session B)

16:30-16:45

Closing Remarks