@misc{UhlmannRizziWenetal., author = {Uhlmann, Max and Rizzi, Tommaso and Wen, Jianan and P{\´e}rez-Bosch Quesada, Emilio and Al Beattie, Bakr and Ochs, Karlheinz and P{\´e}rez, Eduardo and Ostrovskyy, Philip and Carta, Corrado and Wenger, Christian and Kahmen, Gerhard}, title = {LUT-based RRAM model for neural accelerator circuit simulation}, series = {Proceedings of the 18th ACM International Symposium on Nanoscale Architectures}, journal = {Proceedings of the 18th ACM International Symposium on Nanoscale Architectures}, publisher = {ACM}, address = {New York, NY, USA}, doi = {10.1145/3611315.3633273}, pages = {1 -- 6}, abstract = {Neural hardware accelerators have been proven to be energy-efficient when used to solve tasks which can be mapped into an artificial neural network (ANN) structure. Resistive random-access memories (RRAMs) are currently under investigation together with several different memristive devices as promising technologies to build such accelerators combined together with complementary metal-oxide semiconductor (CMOS)-technologies in integrated circuits (ICs). While many research groups are actively developing sophisticated physical-based representations to better understand the underlying phenomena characterizing these devices, not much work has been dedicated to exploit the trade-off between simulation time and accuracy in the definition of low computational demanding models suitable to be used at many abstraction layers. Indeed, the design of complex mixed-signal systems as a neural hardware accelerators requires frequent interaction between the application- and the circuit-level that can be enabled only with the support of accurate and fast-simulating devices' models. In this work, we propose a solution to fill the aforementioned gap with a lookup table (LUT)-based Verilog-A model of IHP's 1-transistor-1-RRAM (1T1R) cell. In addition, the implementation challenges of conveying the communication between the abstract ANN simulation and the circuital analysis are tackled with a design flow for resistive neural hardware accelerators that features a custom Python wrapper. As a demonstration of the proposed design flow and 1T1R model, an ANN for the MNIST handwritten digit recognition task is assessed with the last layer verified in circuit simulation. The obtained recognition confidence intervals show a considerable discrepancy between the purely application-level PyTorch simulation and the proposed design flow which spans across the abstraction layers down to the circuital analysis.}, language = {en} } @misc{UhlmannRizziWenetal., author = {Uhlmann, Max and Rizzi, Tommaso and Wen, Jianan and Quesada, Emilio P{\´e}rez-Bosch and Beattie, Bakr Al and Ochs, Karlheinz and P{\´e}rez, Eduardo and Ostrovskyy, Philip and Carta, Corrado and Wenger, Christian and Kahmen, Gerhard}, title = {End-to-end design flow for resistive neural accelerators}, series = {IEEE transactions on computer-aided design of integrated circuits and systems}, journal = {IEEE transactions on computer-aided design of integrated circuits and systems}, publisher = {Institute of Electrical and Electronics Engineers (IEEE)}, address = {New York}, issn = {0278-0070}, doi = {10.1109/TCAD.2025.3597237}, pages = {1 -- 5}, abstract = {Neural hardware accelerators have demonstrated notable energy efficiency in tackling tasks, which can be adapted to artificial neural network (ANN) structures. Research is currently directed towards leveraging resistive random-access memories (RRAMs) among various memristive devices. In conjunction with complementary metal-oxide semiconductor (CMOS) technologies within integrated circuits (ICs), RRAM devices are used to build such neural accelerators. In this study, we present a neural accelerator hardware design and verification flow, which uses a lookup table (LUT)-based Verilog-A model of IHP's one-transistor-one-RRAM (1T1R) cell. In particular, we address the challenges of interfacing between abstract ANN simulations and circuit analysis by including a tailored Python wrapper into the design process for resistive neural hardware accelerators. To demonstrate our concept, the efficacy of the proposed design flow, we evaluate an ANN for the MNIST handwritten digit recognition task, as well as for the CIFAR-10 image recognition task, with the last layer verified through circuit simulation. Additionally, we implement different versions of a 1T1R model, based on quasi-static measurement data, providing insights on the effect of conductance level spacing and device-to-device variability. The circuit simulations tackle both schematic and physical layout assessment. The resulting recognition accuracies exhibit significant differences between the purely application-level PyTorch simulation and our proposed design flow, highlighting the relevance of circuit-level validation for the design of neural hardware accelerators.}, language = {en} }