%0 Conference Paper
%A Neftci, Emre
%A Yu, Zhenming
%A Leroux, Nathan
%T Training-to-Learn with Memristive Devices
%M FZJ-2024-02778
%D 2023
%X Memristive crossbar arrays are promising non-von Neumann computing technologies to enable real-world, onlinelearning in neural networks. However, their deployment to real-world learning problems is hindered by their non-linearitiesin conductance updates, variation during operation, fabrication mismatch and the realities of gradient descent training. In thiswork, we show that, with a phenomenological model of the device and bi-level optimization, it is possible to pre-train the neuralnetwork to be largely insensitive to such non-idealities on learning tasks. We demonstrate this effect using Model Agnostic Meta Learning (MAML) and a differentiable model of the conductance update on the Omniglot few-shot learning task. Since pre-training is a necessary procedure for any on-line learning scenario at the edge, our results may pave the way towards real-world applications of memristive devices without significant adaption overhead.
%B Neuromorphic Materials, Devices, Circuits and Systems
%C 23 Jan 2023 - 25 Jan 2023, València (Spain)
Y2 23 Jan 2023 - 25 Jan 2023
M2 València, Spain
%F PUB:(DE-HGF)6
%9 Conference Presentation
%R 10.29363/nanoge.neumatdecas.2023.013
%U https://juser.fz-juelich.de/record/1025210