Algorithmic Level Approximate Computing for Machine Learning Classifiers

Younes, Hamoud; Ibrahim, Alì; Rizk, Mostafa; Valle, Maurizio

doi:10.1109/icecs46596.2019.8964974

Cited by 8 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Algorithmic level techniques are divided into two categories: dataoriented and process-oriented. The authors of [17] presented an approach on how to apply these techniques on machine learning algorithms, as shown in Figure 1. The data-oriented category involves modifying the data properties (size and bit-width) to minimize the work-load on the circuit level.…”

Section: Algorithmic Level Approximate Computingmentioning

confidence: 99%

See 1 more Smart Citation

Algorithmic-Level Approximate Tensorial SVM Using High-Level Synthesis on FPGA

et al. 2021

Self Cite

View full text Add to dashboard Cite

Approximate Computing Techniques (ACT) are promising solutions towards the achievement of reduced energy, time latency and hardware size for embedded implementations of machine learning algorithms. In this paper, we present the first FPGA implementation of an approximate tensorial Support Vector Machine (SVM) classifier with algorithmic level ACTs using High-Level Synthesis (HLS). A touch modality classification framework was adopted to validate the effectiveness of the proposed implementation. When compared to exact implementation presented in the state-of-the-art, the proposed implementation achieves a reduction in power consumption by up to 49% with a speedup of 3.2×. Moreover, the hardware resources are reduced by 40% while consuming 82% less energy in classifying an input touch with an accuracy loss less than 5%.

show abstract

Section: Algorithmic Level Approximate Computingmentioning

confidence: 99%

“…Algorithmic level Approximate Computing Techniques: (a) data-oriented; and (b) processing-oriented. Adopted from[17].…”

mentioning

confidence: 99%

Algorithmic-Level Approximate Tensorial SVM Using High-Level Synthesis on FPGA

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…The assessment included the degradation in accuracy to the gain in memory and execution time on Intel i7 CPU. In [8], the studied techniques have been formulated into a general approach that has been tested for the FPGA implementation of two machine learning classifiers. The reported approach has been adopted in this paper where a trade-off between the classifiers performance and quality has been considered.…”

Section: Approximate Knn Blocksmentioning

confidence: 99%

“…The trade-off resulted in our selection for the ACTs presented in the proposed approximate kNN architecture. All the adopted techniques belong to the data-oriented approximate computing category [8]. The adopted ACTs are Dataset Reduction (Downsampling (DS) and Downscaling (DSc)), and Data Format Modification (DFM).…”

Section: Approximate Knn Blocksmentioning

confidence: 99%

See 1 more Smart Citation

An Efficient Selection-Based kNN Architecture for Smart Embedded Hardware Accelerators

Younes¹,

Ibrahim

Rizk

et al. 2021

IEEE Open J. Circuits Syst.

Self Cite

View full text Add to dashboard Cite

K-Nearest Neighbor (kNN) is an efficient algorithm used in many applications e.g. text categorization, data mining, and predictive analysis. Despite having a high computational complexity, kNN is a candidate for hardware acceleration since it is a parallelizable algorithm. This paper presents an efficient novel architecture and implementation for a kNN hardware accelerator targeting modern System-on-Chips (SoCs). The architecture adopts a selection-based sorter dedicated for kNN that outperforms traditional sorters in terms of hardware resources, time latency, and energy efficiency. The kNN architecture has been designed using High-Level Synthesis (HLS) and implemented on the Xilinx Zynqberry platform. Compared to similar state-of-the-art implementations, the proposed kNN provides speedups between 1.4× and 875× with 41% to 94% reductions in energy consumption. To further enhance the proposed architecture, algorithmic-level Approximate Computing Techniques (ACTs) have been applied. The proposed approximate kNN implementation accelerates the classification process by 2.3× with an average reduced area size of 56% for a real-time tactile data processing case study. The approximate kNN consumes 69% less energy with an accuracy loss of less than 3% when compared to the proposed Exact kNN.

show abstract