用强化学习优化存内计算的混合精度训练，提升速度且几乎不损失准确率。
Mixed-Precision Training and Compilation for RRAM-based Computing-in-Memory Accelerators
- 用强化学习自动寻找最佳混合精度配置
- 相比顶尖方案提速最高达2.48倍，精度仅下降0.086%
- 适合研究存内计算与低比特量化加速的工程师
存内计算（CIM）加速器通过在交叉阵列中直接执行矩阵-向量乘法（MVM），有望加速机器学习任务。然而，交叉阵列的输入和单元位宽非常有限，现有大部分CIM编译器不支持低于8位的量化，导致单次MVM需多次计算周期，权重也无法高效存储于单个交叉阵列单元。为此，本文提出一种面向CIM架构的混合精度训练与编译框架。主要挑战在于搜索空间巨大，难以找到最优量化参数。因此，我们引入基于强化学习的策略，自动寻得兼顾延迟与精度的量化配置。实验表明，在最佳情况下，本方法相比现有最先进方案实现高达2.48倍的加速，同时精度损失仅为0.086%。
原文摘要 · Abstract (English)
Computing-in-Memory (CIM) accelerators are a promising solution for accelerating Machine Learning (ML) workloads, as they perform Matrix-Vector Multiplications (MVMs) on crossbar arrays directly in memory. Although the bit widths of the crossbar inputs and cells are very limited, most CIM compilers do not support quantization below 8 bit. As a result, a single MVM requires many compute cycles, and weights cannot be efficiently stored in a single crossbar cell. To address this problem, we propose a mixed-precision training and compilation framework for CIM architectures. The biggest challenge is the massive search space, that makes it difficult to find good quantization parameters. This is why we introduce a reinforcement learning-based strategy to find suitable quantization configurations that balance latency and accuracy. In the best case, our approach achieves up to a 2.48x speedup over existing state-of-the-art solutions, with an accuracy loss of only 0.086 %.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。