Unified matrix-parallel outer-product updates for pretrained deployment and on-device adaptation
Preprint
- 13 May 2026
- preprint
- Published by Springer Nature in Research Square
Abstract
Analog in-memory arrays can reduce the energy of neural inference, but efficient deployment of pretrained networks on hardware is still limited by slow, sequential weight programming and chip-specific mismatch. Weight import and fine-tuning are often handled by different update modes, which increases end-to-end latency and energy in practical workflows. Here we demonstrate a unified outer-product pulse engine in a bidirectional dual-gate charge-trap flash transistor array that executes both low-rank SVD-based weight import and gradient-descent fine-tuning using a single matrix-parallel update primitive. In fabricated arrays, parallel outer-product programming achieves 93% agreement within a 2σ target window and transfers a rank-2 pattern in two update steps. In a closed-loop on-wafer Iris classifier, low-rank import reduces the required fine-tuning iterations from 69 to 3 and cuts training latency by 23×. Scaling analysis projects > 6.4× lower transfer latency, > 64× lower training-update latency, and > 209× lower total training energy for a 128×128 array near 90% accuracy. Together, these results establish an end-to-end, array-native workflow that unifies pretrained weight import and on-device fine-tuning through a single matrix-parallel outer-product primitive.Keywords
All Related Versions
This publication has 37 references indexed in Scilit:
- From commercial to emerging memory: Device and circuit perspectives for neuromorphic computingDevice, 2025
- NeuroTorch: DRAM-free nonvolatile memory-based hybrid training compute-in-memory system simulatorNeurocomputing, 2025
- Polarization Pruning: Reliability Enhancement of Hafnia‐Based Ferroelectric Devices for Memory and Neuromorphic ComputingAdvanced Science, 2024
- Proposition of Adaptive Read Bias: A Solution to Overcome Power and Scaling Limitations in Ferroelectric‐Based Neuromorphic SystemAdvanced Science, 2023
- Si‐Based Dual‐Gate Field‐Effect Transistor Array for Low‐Power On‐Chip Trainable Hardware Neural NetworksAdvanced Intelligent Systems, 2023
- Progress and Challenges for Memtransistors in Neuromorphic Circuits and SystemsAdvanced Materials, 2021
- Bridging Biological and Artificial Neural Networks with Emerging Neuromorphic Devices: Fundamentals, Progress, and ChallengesAdvanced Materials, 2019
- Memristor‐Based Analog Computation and Neural Network Classification with a Dot Product EngineAdvanced Materials, 2018
- Learning representations by back-propagating errorsNature, 1986
- The Approximation of One Matrix by Another of Lower RankPsychometrika, 1936