Chaotic
Systems: Pushing the Frontiers of Classical and Quantum
Dynamics
At
the University of Lethbridge, Dr Keramat Ali is working
on developing new approaches in dynamics, both classical
and quantum, to broaden the efficiency and scope of artificial
recurrent neural networks (RNNs). His focus ranges from
nonlinear dynamics, chaos and noise in information processing
to biometrics and navigation of autonomous agents. Dr
Ali is internationally recognized as an expert in nonlinear
dynamics and his pioneering work in the area of classical
and quantum chaos has brought him to a new dimension of
study in the recurrent neural network field. His current
research endeavors include ascertaining the need for chaos
in achieving meaningful goals, characterizing the role
of chaos and noise in pattern recognition, developing
softwares for biometrics and navigation of autonomous
agents, and designing quantum neural networks.
The
need to study chaotic neural networks is crucial, as
these networks are unavoidable and necessary in real
biological systems. There is a big challenge in physics
though, because there is no persistence in dynamics.
A breakthrough in the area of control and synchronization
of chaos in recurrent neural networks could potentially
transform the development of artificial neural networks
and improve the efficiency of neural network systemization
in robotics applications.
Dr
Ali’s research is pushing the frontiers of classical
and quantum dynamics, but without access to high performance
computational resources he will be unable to continue
with his research breakthroughs. Attempting to systemize
recurrent neural networks and artificial intelligence
requires extensive computation and memory storage. Resources
like MACI are a necessity for computational problems
of this magnitude and to maintain this type of leading
edge research in physics.
ali@hg.uleth.ca
Selected
Publications
M.
Andrecut and M. K Ali. Example of Robust Chaos in
a Smooth Map, Europhysics
Letters, in press, 2001.
M.
Andrecut and M. K Ali. Chaos in a Simple Boolean
Network, International Journal of Modern Physics,
B 14, in press, 2001.
M.
Andrecut and M. K Ali. A Simple Neural Network Approach
to Invariant Image Recognition, Modern Physics
Letters B, in press, 2001.
Z.
Tan and M. K Ali. Associative Memory using Synchronization
in a Chaotic Neural Network, International Journal
of Modern Physics, C 12, No 1, in press, 2001.
Z.
Tan and M. K. Ali. Pattern Recognition with Stochastic
Resonance in a Generic Neural Network, International
Journal of Modern Physics, C 11, No. 8, in press,
2001.
·A.
Potapov and M. K Ali. Chaotic Neural Control,
Physical Review E 63, in press 2001.
Potapov
and M.K. Ali. Robust Chaos in Neural Networks,
Physics Letters A 277, 310322, 2000.
M.
Andrecut and M. K Ali. Selforganizing Neural Network
Model of the SensoryMotor Mechanism, International
Journal of Modern Physics, B 14, 18151824, 2000.
