Efficient Computing for Robotics and AI

In this talk, we will describe how the joint algorithm and hardware design can be used to reduce energy consumption while delivering real-time and robust performance for applications including deep learning, computer vision, autonomous navigation/exploration, and video/image processing.

This talk has a greater emphasis on our robotics work compared to "Efficient Computing for AI and Robotics" which focuses more on our AI work.

Works highlighted in this talk include

* Y.H. Chen, T. Krishna, J. Emer, V. Sze, “EyerissAn EnergyEfficient Reconfigurable Accelerator for Deep Convolutional Neural Networks,” IEEE Journal of SolidState Circuits (JSSC), ISSCC Special Issue, Vol. 52, No. 1, pp. 127138, January 2017. [Project website ]

* A. Suleiman*, Y.H. Chen*, J. Emer, V. Sze, “Towards Closing the Energy Gap Between HOG and CNN Features for Embedded Vision,” IEEE International Symposium of Circuits and Systems (ISCAS), Invited Paper, May 2017.

* T.J. Yang, Y.H. Chen, V. Sze, “Designing EnergyEfficient Convolutional Neural Networks using EnergyAware Pruning,” IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 2017.

* T.J. Yang, A. Howard, B. Chen, X. Zhang, A. Go, M. Sandler, V. Sze, H. Adam, “NetAdaptPlatformAware Neural Network Adaptation for Mobile Applications,” European Conference on Computer Vision (ECCV), September 2018.

*D. Wofk*, F. Ma*, T.J. Yang, S. Karaman, V. Sze, “FastDepthFast Monocular Depth Estimation on Embedded Systems,” IEEE International Conference on Robotics and Automation (ICRA), May 2019.

*Y.H. Chen, T.J Yang, J. Emer, V. Sze, “Eyeriss v2A Flexible Accelerator for Emerging Deep Neural Networks on Mobile Devices,” to appear in IEEE Journal on Emerging and Selected Topics in Circuits and Systems (JETCAS), June 2019.

* A. Suleiman, Z. Zhang, L. Carlone, S. Karaman, V. Sze, “NavionA 2mW Fully Integrated RealTime VisualInertial Odometry Accelerator for Autonomous Navigation of Nano Drones,” IEEE Journal of SolidState Circuits (JSSC), VLSI Symposia Special Issue, Vol. 54, No. 4, pp. 11061119, April 2019. [Project website ]

* Z. Zhang, T. Henderson, S. Karaman, V. Sze, “FSMIFast computation of Shannon Mutual Information for informationtheoretic mapping,” arXiv, May 2019.

* P. Z. X. Li*, Z. Zhang*, S. Karaman, V. Sze, “Highthroughput Computation of Shannon Mutual Information on Chip,” RoboticsScience and Systems (RSS), June 2019.

* J. Noraky, V. Sze, “Low Power Depth Estimation of Rigid Objects for TimeofFlight Imaging,” to appear in IEEE Transactions on Circuits and Systems for Video Technology (TCSVT).