Article Dans Une Revue Advanced Functional Materials Année : 2025

All‐in‐One Analog AI Hardware: On‐Chip Training and Inference with Conductive‐Metal‐Oxide/HfOx ReRAM Devices

Donato Francesco Falcone
Victoria Clerico
Wooseok Choi
Tommaso Stecconi
Matteo Galetta
Bert Jan Offrein
Valeria Bragaglia

Résumé

Analog in-memory computing is an emerging paradigm designed to efficiently accelerate deep neural network workloads. Recent advancements have focused on either inference or training acceleration. However, a unified analog in-memory technology platform-capable of on-chip training, weight retention, and long-term inference acceleration-has yet to be reported. This work presents an all-in-one analog AI accelerator, combining these capabilities to enable energy-efficient, continuously adaptable AI systems. The platform leverages an array of analog filamentary conductive-metal-oxide (CMO)/HfOx resistive switching memory cells (ReRAM) integrated into the back-end-of-line (BEOL). The array demonstrates reliable resistive switching with voltage amplitudes below 1.5 V, compatible with advanced technology nodes. The array's multi-bit capability (over 32 stable states) and low programming noise (down to 10 nS) enable a nearly ideal weight transfer process, more than an order of magnitude better than other memristive technologies. Inference performance is validated through matrix-vector multiplication simulations on a 64 × 64 array, achieving a root-mean-square error improvement by a factor of 20 at 1 s and 3 at 10 years after programming, compared to state-of-the-art. Training accuracy closely matching the software equivalent is achieved across different datasets. The CMO/HfOx ReRAM technology lays the foundation for efficient analog systems accelerating both inference and training in deep neural networks.

Fichier principal
Vignette du fichier
Adv Funct Materials - 2025 - Falcone - All‐in‐One Analog AI Hardware On‐Chip Training and Inference with.pdf (7.38 Mo) Télécharger le fichier
Origine Publication financée par une institution
Licence

Dates et versions

hal-05148002 , version 1 (07-07-2025)

Licence

Identifiants

Citer

Donato Francesco Falcone, Victoria Clerico, Wooseok Choi, Tommaso Stecconi, Folkert Horst, et al.. All‐in‐One Analog AI Hardware: On‐Chip Training and Inference with Conductive‐Metal‐Oxide/HfOx ReRAM Devices. Advanced Functional Materials, 2025, pp.2504688. ⟨10.1002/adfm.202504688⟩. ⟨hal-05148002⟩
86 Consultations
117 Téléchargements

Altmetric

Partager

  • More