A High Energy Efficient Reconfigurable Hybrid Neural Network Processor for Deep Learning Applications

• Published in: IEEE journal of solid-state circuits 2018-04, Vol.53 (4), p.968-982
• Main Authors: Yin, Shouyi, Ouyang, Peng, Tang, Shibin, Tu, Fengbin, Li, Xiudong, Zheng, Shixuan, Lu, Tianyi, Gu, Jiangyuan, Liu, Leibo, Wei, Shaojun
• Format: Article
• Language: English
• Subjects: Acceleration; Arrays; Artificial neural networks; Computation; Computer memory; Deep learning; Energy consumption; Energy efficiency; Energy measurement; hybrid neural networks (hybrid-NNs); memory banking; Microprocessors; Neural networks; Power efficiency; reconfigurable computing; Reconfiguration; resource partitioning; Speech recognition; Throughput
• ISSN: 0018-9200; 1558-173X
• DOI: 10.1109/JSSC.2017.2778281

Hybrid neural networks (hybrid-NNs) have been widely used and brought new challenges to NN processors. Thinker is an energy efficient reconfigurable hybrid-NN processor fabricated in 65-nm technology. To achieve high energy efficiency, three optimization techniques are proposed. First, each processing element (PE) supports bit-width adaptive computing to meet various bit-widths of neural layers, which raises computing throughput by 91% and improves energy efficiency by 1.93 \times on average. Second, PE array supports on-demand array partitioning and reconfiguration for processing different NNs in parallel, which results in 13.7% improvement of PE utilization and improves energy efficiency by 1.11 \times . Third, a fused data pattern-based multi-bank memory system is designed to exploit data reuse and guarantee parallel data access, which improves computing throughput and energy efficiency by 1.11 \times and 1.17 \times , respectively. Measurement results show that this processor achieves 5.09-TOPS/W energy efficiency at most.