analog AI at IBM

inspiring review Analog architectures for neural network acceleration based on non-volatile memory

the challenges are the overhead imposed by peripheral circuitry and non-ideal properties of neuron cells.

https://www.ibm.com/blogs/research/2020/12/iedm2020-memory-analog-ai/

Run an AI algorithm where computation and storage coexist on a single chip

IBM - Analog AI

• 2020 Unassisted True Analog Neural Network Training Chip

• dec 2020 Precision of synaptic weights programmed in phase-change memory devices for deep learning inference Date of Conference: 12-18 Dec. 2020

• may 2020 Accurate deep neural network inference using computational phase-change memory - Vinay Joshi

Analog In-Memory Computing using Resistive Processing Unit (RPU) has been proposed for Neural Network (NN) training.

• video SDC2020: Analog Memory-based Techniques for Accelerating Deep Neural Networks

• presentation 2019 Accelerating Deep Learning with Analog Memory - A Device, Circuit and Systems Approach and the full article

sept 2020 - Analog Memory-Based Techniques for Accelerating Deep Neural Networks

Authors: Tayfun Gokmen, Wilfried Haensch

• The Next Generation of Deep Learning Hardware: Analog Computing
  This paper explores the current state of neuromorphic deep learning architectures in silicon CMOS technology. By WILFRIED HAENSCH

news

https://www.ibm.com/blogs/research/2020/12/iedm2020-memory-analog-ai/

https://www.ibm.com/blogs/research/2018/07/capacitor-ai-accelerators/