20.500.12556/RUL-135589 An approximate GEMM unit for energy-efficient object detection Edge computing brings artificial intelligence algorithms and graphics processing units closer to data sources, making autonomy and energy-efficient processing vital for their design. Approximate computing has emerged as a popular strategy for energy-efficient circuit design, where the challenge is to achieve the best tradeoff between design efficiency and accuracy. The essential operation in artificial intelligence algorithms is the general matrix multiplication (GEMM) operation comprised of matrix multiplication and accumulation. This paper presents an approximate general matrix multiplication (AGEMM) unit that employs approximate multipliers to perform matrix–matrix operations on four-by-four matrices given in sixteen-bit signed fixed-point format. The synthesis of the proposed AGEMM unit to the 45 nm Nangate Open Cell Library revealed that it consumed only up to 36% of the area and 25% of the energy required by the exact general matrix multiplication unit. The AGEMM unit is ideally suited to convolutional neural networks, which can adapt to the error induced in the computation. We evaluated the AGEMM units’ usability for honeybee detection with the YOLOv4-tiny convolutional neural network. The results implied that we can deploy the AGEMM units in convolutional neural networks without noticeable performance degradation. Moreover, the AGEMM unit’s employment can lead to more area- and energy-efficient convolutional neural network processing, which in turn could prolong sensors’ and edge nodes’ autonomy. approximate computing logarithmic multiplier computer arithmetic energy-efficient processing tensor core approximate general matrix multiplication GEMM matrix core approximate multipliers convolutional neural networks object detection YOLOv4-tiny honeybee detection približno računanje logaritemski množilnik računalniška aritmetika energijsko učinkovito procesiranje tensorsko jedro true false true Angleški jezik Slovenski jezik Članek v reviji 2022-03-21 14:13:35 2022-03-21 14:13:37 2022-09-14 04:14:07 0000-00-00 00:00:00 2021 0 0 19 str. iss. 12, art. 4195 Vol. 21 Jun. 2021 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004 1424-8220 10.3390/s21124195 67593475 RAZ_Pilipovic_Ratko_2021.pdf RAZ_Pilipovic_Ratko_2021.pdf 1 00FF73424D011A10FDACAF3E3E755D83 0851607a326df9469cf9d4871266ebf8db4540b776466096cdcacc90d1f63ad8 ea1b98cf-a918-11ec-8aca-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=154748 https://www.mdpi.com/1424-8220/21/12/4195 1 b860a529-a918-11ec-8aca-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=154747 Fakulteta za računalništvo in informatiko Biotehniška fakulteta 0 0 0