ReckOn: A 28nm Sub-mm2 Task-Agnostic Spiking Recurrent Neural Network Processor Enabling On-Chip Learning over Second-Long Timescales

📋 论文概要

该论文提出了一种名为ReckOn的28nm亚平方毫米任务无关脉冲递归神经网络处理器，支持片上学习，解决了边缘设备在数据分布变化下的自适应问题。通过创新的学习算法和硬件设计，实现了在长时间尺度上的在线学习，同时满足严格的功耗和面积约束。

💡 主要创新点

• 提出了一种任务无关的脉冲递归神经网络处理器架构，支持片上学习，无需外部干预即可适应数据分布变化。
• 开发了与时间深度无关的学习算法，避免了传统训练算法中内存需求随数据时间深度线性增长的问题。
• 在28nm工艺下实现了亚平方毫米的芯片面积，展示了极低的功耗和面积开销，适合边缘部署。

🏷 关键词

📄 原文摘要

The robustness of autonomous inference-only devices deployed in the real world is limited by data distribution changes induced by different users, environments, and task requirements. This challenge calls for the development of edge devices with an alwayson adaptation to their target ecosystems. However, the memory requirements of conventional neural-network training algorithms scale with the temporal depth of the data being processed, which is not compatible with the constrained power and area budgets at the edge. For this reason, previous works demonstrating end-to-end on-chip learning without external memory were restricted to the processing of static data such as images [1-4], or to instantaneous decisions involving no memory of the past, e.g. obstacle avoidance in mobile robots [5]. The ability to learn short-to-long-term temporal dependencies on-chip is a missing enabler for robust autonomous edge devices

👥 作者与机构

Charlotte Frenkel, Giacomo Indiveri

University of Zurich and ETH Zurich, Zurich, Switzerland