Hardware architecture of the protein processing associative memory and the effects of dimensionality and quantisation on performance

Omer Qadir*, Alex Lenz, Gianluca Tempesti, Jon Timmis, Tony Pipe, Andy Tyrrell

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (SciVal)

Abstract

The Protein Processor Associative Memory (PPAM) is a novel hardware architecture for a distributed, decentralised, robust and scalable, bidirectional, hetero-associative memory, that can adapt online to changes in the training data. The PPAM uses the location of data in memory to identify relationships and is therefore fundamentally different from traditional processing methods that tend to use arithmetic operations to perform computation. This paper presents the hardware architecture and details a sample digital logic implementation with an analysis of the implications of using existing techniques for such hardware architectures. It also presents the results of implementing the PPAM for a robotic application that involves learning the forward and inverse kinematics. The results show that, contrary to most other techniques, the PPAM benefits from higher dimensionality of data, and that quantisation intervals are crucial to the performance of the PPAM.

Original languageEnglish
Pages (from-to)245-274
Number of pages30
JournalGenetic Programming and Evolvable Machines
Volume15
Issue number3
DOIs
Publication statusPublished - Sept 2014

Keywords

  • Associative memory
  • BERT2
  • Dimensionality
  • FPGA
  • Inverse kinematics
  • Non-standard computation
  • PPAM
  • Protein processing
  • Quantisation

Fingerprint

Dive into the research topics of 'Hardware architecture of the protein processing associative memory and the effects of dimensionality and quantisation on performance'. Together they form a unique fingerprint.

Cite this