Implementation of offset corrected AGAD algorithm on 130-nm CMOS technology–based RRAM array for analog neural network training

Jimin Lee; Paul Michael Solomon; Nanbo Gong; Omobayode I. Fagbohungbe; Russel Wilson; Takashi Ando; Seyoung Kim

doi:10.1126/sciadv.aee7134

AI & machine learningEfficiency & hardware

Implementation of offset corrected AGAD algorithm on 130-nm CMOS technology–based RRAM array for analog neural network training

Jimin Lee, Paul Michael Solomon, Nanbo Gong, Omobayode I. Fagbohungbe, Russel Wilson, Takashi Ando, Seyoung Kim

Peer-reviewed journalBold claims, read criticallyClaims a big step

In the authors' words

This journal does not let us republish the full abstract. Here are the two sentences Pipette selected, quoted from it. Read the rest at the publisher.

Main result

We present the first implementation of the analog gradient accumulation with dynamic reference (AGAD), reported as the most advanced and highest-performing version of the TT (Tiki-Taka) algorithm, on an HfO 2 -based resistive random-access memory (RRAM) array for analog neural network training.

Read the paper

Appeared: Sunday, September 27. Science Advances. Peer-reviewed journal.

DOI: 10.1126/sciadv.aee7134

Implementation of offset corrected AGAD algorithm on 130-nm CMOS technology–based RRAM array for analog neural network training

In the authors' words

More on Efficiency & hardware that day

Multiplexed deep visual proteomics resolves spatial heterogeneity and rare endocrine states in human pancreatic islets

Deep learning extracts MoA-specific signatures from high-throughput images of chemically and genetically perturbed Corynebacteria

Discovering how ice crystals grow using neural ODEs and symbolic regression

Optical microscopy predictions of focal recurrence in glioblastoma

A structure-informed deep learning framework for modeling TCR-peptide-HLA interactions