# Harnessing Nonidealities in Analog In-Memory Computing Circuits: A Physical Modeling Approach for Neuromorphic Systems

arXiv: https://arxiv.org/abs/2412.09010 · HTML: https://mlsanity.com/paper/2412.09010/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 49 references that are in the map.
- First posted on arXiv 12 Dec 2024.

## Builds on (most cited first, 18 of 18)

- [In-Datacenter Performance Analysis of a Tensor Processing Unit](https://mlsanity.com/paper/1704.04760/) (2017)
- [Explaining Neural Scaling Laws](https://mlsanity.com/paper/2102.06701/) (2021)
- [SuperSpike: Supervised learning in multi-layer spiking neural networks](https://mlsanity.com/paper/1705.11146/) (2017)
- [Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems](https://mlsanity.com/paper/2104.13386/) (2021)
- [Accurate and efficient time-domain classification with adaptive spiking recurrent neural networks](https://mlsanity.com/paper/2103.12593/) (2021)
- [Machine Learning at the Network Edge: A Survey](https://mlsanity.com/paper/1908.00080/) (2019)
- [PUMA: A Programmable Ultra-efficient Memristor-based Accelerator for Machine Learning Inference](https://mlsanity.com/paper/1901.10351/) (2019)
- [Event-Based Backpropagation can compute Exact Gradients for Spiking Neural Networks](https://mlsanity.com/paper/2009.08378/) (2020)
- [Fast and energy-efficient neuromorphic deep learning with first-spike times](https://mlsanity.com/paper/1912.11443/) (2019)
- [Temporal-Coded Deep Spiking Neural Network with Easy Training and Robust Performance](https://mlsanity.com/paper/1909.10837/) (2019)
- [Hardware-aware training for large-scale and diverse deep learning inference workloads using in-memory computing-based accelerators](https://mlsanity.com/paper/2302.08469/) (2023)
- [Backpropagation-free Training of Deep Physical Neural Networks](https://mlsanity.com/paper/2304.11042/) (2023)
- [Neuromorphic Deep Learning Machines](https://mlsanity.com/paper/1612.05596/) (2016)
- [Surrogate gradients for analog neuromorphic computing](https://mlsanity.com/paper/2006.07239/) (2020)
- [Elucidating the theoretical underpinnings of surrogate gradient learning in spiking neural networks](https://mlsanity.com/paper/2404.14964/) (2024)
- [A Review of SRAM-based Compute-in-Memory Circuits](https://mlsanity.com/paper/2411.06079/) (2024)
- [Sparse-firing regularization methods for spiking neural networks with time-to-first spike coding](https://mlsanity.com/paper/2307.13007/) (2023)
- [3D-aCortex: An Ultra-Compact Energy-Efficient Neurocomputing Platform Based on Commercial 3D-NAND Flash Memories](https://mlsanity.com/paper/1908.02472/) (2019)

## Built on by (most cited first, 1 of 1)

- [Scalable Training of Continuous-Time Spiking Neural Networks with Differentiable Spike-Time Discretization](https://mlsanity.com/paper/2607.14672/) (2026)

## Abstract (arXiv)

Large-scale deep learning models are increasingly constrained by their immense energy consumption, limiting their scalability and applicability for edge intelligence. In-memory computing (IMC) offers a promising solution by addressing the von Neumann bottleneck inherent in traditional deep learning accelerators, significantly reducing energy consumption. However, the analog nature of IMC introduces hardware nonidealities that degrade model performance and reliability. This paper presents a novel approach to directly train physical models of IMC, formulated as ordinary-differential-equation (ODE)-based physical neural networks (PNNs). To enable the training of large-scale networks, we propose a technique called differentiable spike-time discretization (DSTD), which reduces the computational cost of ODE-based PNNs by up to 20 times in speed and 100 times in memory. We demonstrate that such large-scale networks enhance the learning performance by exploiting hardware nonidealities on the CIFAR-10 dataset. The proposed bottom-up methodology is validated through the post-layout SPICE simulations on the IMC circuit with nonideal characteristics using the sky130 process. The proposed PNN approach reduces the discrepancy between the model behavior and circuit dynamics by at least an order of magnitude. This work paves the way for leveraging nonideal physical devices, such as non-volatile resistive memories, for energy-efficient deep learning applications.