AHaH Computing–From Metastable Switches to Attractors to Machine Learning
Authors:
Open Access
- 10 February 2014
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 9 (2) , e85175
Abstract
Modern computing architecture based on the separation of memory and processing leads to a well known problem called the von Neumann bottleneck, a restrictive limit on the data bandwidth between CPU and RAM. This paper introduces a new approach to computing we call AHaH computing where memory and processing are combined. The idea is based on the attractor dynamics of volatile dissipative electronics inspired by biological systems, presenting an attractive alternative architecture that is able to adapt, self-repair, and learn from interactions with the environment. We envision that both von Neumann and AHaH computing architectures will operate together on the same machine, but that the AHaH computing processor may reduce the power consumption and processing time for certain adaptive learning tasks by orders of magnitude. The paper begins by drawing a connection between the properties of volatility, thermodynamics, and Anti-Hebbian and Hebbian (AHaH) plasticity. We show how AHaH synaptic plasticity leads to attractor states that extract the independent components of applied data streams and how they form a computationally complete set of logic functions. After introducing a general memristive device model based on collections of metastable switches, we show how adaptive synaptic weights can be formed from differential pairs of incremental memristors. We also disclose how arrays of synaptic weights can be used to build a neural node circuit operating AHaH plasticity. By configuring the attractor states of the AHaH node in different ways, high level machine learning functions are demonstrated. This includes unsupervised clustering, supervised and unsupervised classification, complex signal prediction, unsupervised robotic actuation and combinatorial optimization of procedures–all key capabilities of biological nervous systems and modern machine learning algorithms with real world application.Keywords
This publication has 97 references indexed in Scilit:
- Arabidopsis plants perform arithmetic division to prevent starvation at nighteLife, 2013
- Nanoscale electronic synapses using phase change devicesACM Journal on Emerging Technologies in Computing Systems, 2013
- Ultrafast Synaptic Events in a Chalcogenide MemristorScientific Reports, 2013
- Memristor-based neural networksJournal of Physics D: Applied Physics, 2013
- Machine learning: a review of classification and combining techniquesArtificial Intelligence Review, 2006
- Simple model of spiking neuronsIEEE Transactions on Neural Networks, 2003
- Forecasting with artificial neural networks:International Journal of Forecasting, 1998
- Constructal-theory network of conducting paths for cooling a heat generating volumeInternational Journal of Heat and Mass Transfer, 1997
- Emergent attractors and the law of maximum entropy production: Foundations to a theory of general evolutionSystems Research, 1989
- The chemical basis of morphogenesisPhilosophical Transactions of the Royal Society of London. B, Biological Sciences, 1952