On Quantification of Accuracy Loss in Approximate Computing
نویسندگان
چکیده
Emerging applications such as R(ecognition), M(ining), and S(ynthesis) suit themselves well to approximate computing due to their intrinsic noise tolerance. RMS applications process massive, yet noisy and redundant data by probabilistic, often iterative, algorithms. Usually the solution space has many more elements than one, rendering a range of application outputs valid, as opposed to a single golden value. A critical step in translating this intrinsic noise tolerance to energy efficiency is quantification of approximation-induced accuracy loss using application-specific metrics. This article covers pitfalls and fallacies in the development and deployment of accuracy metrics.
منابع مشابه
Imprecise Minority-Based Full Adder for Approximate Computing Using CNFETs
Nowadays, the portable multimedia electronic devices, which employ signal-processing modules, require power aware structures more than ever. For the applications associating with human senses, approximate arithmetic circuits can be considered to improve performance and power efficiency. On the other hand, scaling has led to some limitations in performance of nanoscale circuits. According...
متن کاملA Hybrid Neural Network Approach for Kinematic Modeling of a Novel 6-UPS Parallel Human-Like Mastication Robot
Introduction we aimed to introduce a 6-universal-prismatic-spherical (UPS) parallel mechanism for the human jaw motion and theoretically evaluate its kinematic problem. We proposed a strategy to provide a fast and accurate solution to the kinematic problem. The proposed strategy could accelerate the process of solution-finding for the direct kinematic problem by reducing the number of required ...
متن کاملResistive Memory for Approximate Program Acceleration
The Internet of Things significantly increases the amount of data generated that strains the processing capability of current computing systems. Approximate computing can accelerate the computation and dramatically reduce the energy consumption with controllable accuracy loss. In this paper, we propose a Resistive Associative Unit, called RAU, which approximates computation alongside processing...
متن کاملApproximate Solution of Sensitivity Matrix of Required Velocity Using Piecewise Linear Gravity Assumption
In this paper, an approximate solution of sensitivity matrix of required velocity with final velocity constraint is derived using a piecewise linear gravity assumption. The total flight time is also fixed for the problem. Simulation results show the accuracy of the method. Increasing the midway points for linearization, increases the accuracy of the solution, which this, in turn, depends on the...
متن کاملPothole Detection by Soft Computing
Subject- Potholes on roads are regarded as serious problems in the transportation domain and ignoring them leads to the increase of accidents, traffic, vehicle fuel consumption and waste of time and energy. As a result, pothole detection has attracted researchers’ attention and different methods have been presented for it up to now. Background- The major part of previous research is based on i...
متن کامل