Ortner Thomas, Petschenig Horst, Vasilopoulos Athanasios, Renner Roland, Brglez Špela, Limbacher Thomas, Piñero Enrique, Linares-Barranco Alejandro, Pantazi Angeliki, Legenstein Robert
IBM Research Europe - Zurich, Rüschlikon, Switzerland.
Institute of Machine Learning and Neural Computation, Graz University of Technology, Graz, Austria.
Nat Commun. 2025 Feb 1;16(1):1243. doi: 10.1038/s41467-025-56345-4.
There is a growing demand for low-power, autonomously learning artificial intelligence (AI) systems that can be applied at the edge and rapidly adapt to the specific situation at deployment site. However, current AI models struggle in such scenarios, often requiring extensive fine-tuning, computational resources, and data. In contrast, humans can effortlessly adjust to new tasks by transferring knowledge from related ones. The concept of learning-to-learn (L2L) mimics this process and enables AI models to rapidly adapt with only little computational effort and data. In-memory computing neuromorphic hardware (NMHW) is inspired by the brain's operating principles and mimics its physical co-location of memory and compute. In this work, we pair L2L with in-memory computing NMHW based on phase-change memory devices to build efficient AI models that can rapidly adapt to new tasks. We demonstrate the versatility of our approach in two scenarios: a convolutional neural network performing image classification and a biologically-inspired spiking neural network generating motor commands for a real robotic arm. Both models rapidly learn with few parameter updates. Deployed on the NMHW, they perform on-par with their software equivalents. Moreover, meta-training of these models can be performed in software with high-precision, alleviating the need for accurate hardware models.
对低功耗、自主学习的人工智能（AI）系统的需求日益增长，这类系统可应用于边缘设备，并能快速适应部署现场的具体情况。然而，当前的AI模型在这种场景下表现不佳，通常需要大量的微调、计算资源和数据。相比之下，人类可以通过从相关任务中转移知识轻松适应新任务。学习学习（L2L）的概念模仿了这一过程，使AI模型只需很少的计算量和数据就能快速适应。内存计算神经形态硬件（NMHW）受大脑工作原理的启发，模仿了其内存和计算在物理上的共置。在这项工作中，我们将L2L与基于相变存储器件的内存计算NMHW相结合，构建能够快速适应新任务的高效AI模型。我们在两种场景中展示了我们方法的通用性：一个执行图像分类的卷积神经网络和一个为真实机器人手臂生成运动命令的受生物启发的脉冲神经网络。这两种模型都只需很少的参数更新就能快速学习。部署在NMHW上时，它们的性能与软件等效模型相当。此外，这些模型的元训练可以在软件中高精度地进行，从而减少了对精确硬件模型的需求。
