FIXAR：一个具有Quantization感知训练和自适应学习的定点深度强化学习平台
FIXAR: A Fixed-Point Deep Reinforcement Learning Platform with Quantization-Aware Training and Adaptive Parallelism
摘要：
在本文中，我们提出了一个名为FIXAR的深度强化学习平台，该平台首次采用了固定点数据类型和算术单元，并采用了软硬件协同设计的方法。从32位定点数据开始，量化感知训练（QAT）根据激活范围降低其数据精度，并执行再训练以最大限度地减少奖励降级。FIXAR提出了由可配置处理元件组成的自适应阵列处理核心，以支持高吞吐量推理和训练的层内并行和批内并行。最后，在Xilinx U 50上实现了FIXAR，达到了每秒25293.3次推理（IPS）的训练吞吐量和2638.0 IPS/W的加速器效率，比CPU-GPU平台快2.7倍，节能15.4倍，而精度没有任何下降。
英文摘要：
In this paper, we present a deep reinforcement learning platform named FIXAR which employs fixed-point data types and arithmetic units for the first time using a SW/HW co-design approach. Starting from 32-bit fixed-point data, Quantization-Aware Training (QAT) reduces its data precision based on the range of activations and performs retraining to minimize the reward degradation. FIXAR proposes the adaptive array processing core composed of configurable processing elements to support both intra-layer parallelism and intra-batch parallelism for high-throughput inference and training. Finally, FIXAR was implemented on Xilinx U50 and achieves 25293.3 inferences per second (IPS) training throughput and 2638.0 IPS/W accelerator efficiency, which is 2.7 times faster and 15.4 times more energy efficient than those of the CPU-GPU platform without any accuracy degradation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
