Title: Associative%20Memory%20by%20Recurrent%20Neural%20Networks%20with%20Delay%20Elements
1Associative Memory by Recurrent Neural Networks
with Delay Elements
- Seiji MIYOSHI Hiro-Fumi YANAI Masato
OKADA
Kobe City College of Tech. Ibaraki
Univ. RIKEN BSI , ERATO KDB
JAPAN JAPAN
JAPAN
miyoshi_at_kobe-kosen.ac.jp www.kobe-kosen.ac.jp/miy
oshi/
2Background
- Synapses of real neural systems seem to have
delays.
- It is very important to analyze associative
memory model with delayed synapses.
- Computer simulation is powerful method.
There is a Limit on the number of neurons.
However,
Simulating network with large delay steps is
realistically impossible.
- Theoretical and analytical approach is
indispensable to research on delayed networks.
- Yanai-Kim theory by using Statistical
Neurodynamics
Good Agreement with computer simulation
Computational Complexity is O(L4t)
3Objective
- To derive macroscopic steady state equations by
using discrete Fourier transformation
- To discuss storage capacity quantitatively even
for a large L limit (L length of delay)
4Recurrent Neural Network with Delay Elements
Model
5Model
6Macrodynamical Equations by Statistical
NeurodynamicsYanai Kim(1995) Miyoshi, Yanai
Okada(2002)
7Initial Condition of the Network
- One Step Set Initial Condition
- Only the states of neurons are set explicitly.
- The states of delay elements are set to be zero.
- All Steps Set Initial Condition
- The states of all neurons and all delay elements
are set to be close to the stored pattern
sequences. - If they are set to be the stored pattern
sequences themselves - Optimum Initial Condition
8Dynamical Behaviors of Recall Process
All Steps Set Intial Condition Loading
ratea0.5 Length of delay L3
Theory
Simulation(N2000)
9Dynamical Behaviors of Recall Process
All Steps Set Intial Condition Loading
ratea0.5 Length of delay L2
Theory
Simulation(N2000)
10Loading rates a - Steady State Overlaps m
Theory
Simulation(N500)
11Length of delay L - Critical Loading Rate aC
12Macrodynamical Equations by Statistical
NeurodynamicsYanai Kim(1995) Miyoshi, Yanai
Okada(2002)
- Good Agreement with Computer Simulation
- Computational Complexity is O(L4t)
13Macroscopic Steady State Equations
- Accounting for Steady State
- Parallel Symmetry in terms of Time Steps
- Discrete Fourier Transformation
14Loading rates a - Steady State Overlaps m
15Loading rates a - Steady State Overlaps m
Theory
Simulation(N500)
16Loading rate a - Steady State Overlap
17Storage Capacity of Delayed Network
Storage Capacity 0.195 L
18Conclusions
- Yanai-Kim theory (macrodynamical equations for
delayed network) is re-derived.
? Computational Complexity is O(L4t)
? Intractable to discuss macroscopic
properties in a large L limit
- Steady state equations are derived by using
discrete Fourier transformation.
? Computational complexity does not formally
depend on L
? Phase transition points agree with those
under the optimum initial conditions, that is,
the Storage Capacities !
- Storage capacity is 0.195 L in a large L limit.