Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator
ACM Transactions on Modeling and Computer Simulation (TOMACS) - Special issue on uniform random number generation
Reliable low-power digital signal processing via reduced precision redundancy
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Variable latency speculative addition: a new paradigm for arithmetic circuit design
Proceedings of the conference on Design, automation and test in Europe
Proceedings of the 14th ACM/IEEE international symposium on Low power electronics and design
Rodinia: A benchmark suite for heterogeneous computing
IISWC '09 Proceedings of the 2009 IEEE International Symposium on Workload Characterization (IISWC)
Slack redistribution for graceful degradation under voltage overscaling
Proceedings of the 2010 Asia and South Pacific Design Automation Conference
Exploring the fidelity-efficiency design space using imprecise arithmetic
Proceedings of the 16th Asia and South Pacific Design Automation Conference
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
Improving energy gains of inexact DSP hardware through reciprocative error compensation
Proceedings of the 50th Annual Design Automation Conference
An energy efficient approximate adder with carry skip for error resilient neuromorphic VLSI systems
Proceedings of the International Conference on Computer-Aided Design
Hi-index | 0.00 |
Recent studies have demonstrated the potential for reducing energy consumption in integrated circuits by allowing errors during computation. While most proposed techniques for achieving this rely on voltage overscaling (VOS), this paper shows that Imprecise Hardware (IHW) with design-time structural parameters can achieve orthogonal energy-quality tradeoffs. Two IHW adders are improved and two IHW multipliers are introduced in this paper. In addition, a simulation-free error estimation technique is proposed to rapidly and accurately estimate the impact of IHW on output quality. Finally, a quality-aware energy minimization methodology is presented. To validate this methodology, experiments are conducted on two computational kernels: DOT-PRODUCT and L2-NORM -- used in three applications -- Leukocyte Tracker, SVM classification and K-means clustering. Results show that the Hellinger distance between estimated and simulated error distribution is within 0.05 and that the methodology enables designers to explore energy-quality tradeoffs with significant reduction in simulation complexity.