Learning-to-learn enables rapid learning with phase-change memory-based in-memory computing
This work addresses the need for efficient, adaptive AI at the edge, though it is incremental as it combines existing L2L and in-memory computing concepts.
The authors tackled the problem of enabling low-power AI systems to rapidly adapt to new tasks with minimal data and computational resources by pairing learning-to-learn (L2L) with phase-change memory-based in-memory computing hardware. They demonstrated that models for image classification and robotic motor control achieve performance on-par with software equivalents while requiring few parameter updates.
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.