[Comp-neuro] Topic: Special Session in ICONIP 2012. "Non-stationary Time Series Processing in Computational Neuroscience"

Topic: Special Session in ICONIP 2012. “Non-stationary Time Series Processing in Computational Neuroscience”

 

Submissions are welcome to the special session entitled “Non-stationary Time Series Processing in Computational Neuroscience”, at the 19^th International Conference on Neural Information Processing.

Doha, November 12-15, 2012

http://www.iconip2012.org/plenary_speakers.htm

Deadline for the submission is 1^th of June.           

 

Scope of the special session:

 

One of the most fundamental and yet unsolved problems in neuroscience is the understanding of the dynamics underlying the immediate and effortless perceptual processes. For instance, the reliable identification of speaker’s age, sex, identity or musical instruments in a noisy environment requires the on-line disentanglement of sophisticated temporal patterns of sounds, characterized by hundreds of variables (frequencies) which are separated by a fraction of a millisecond. In other words, the enormous complexity of the human cognitive landscape requires the existence of an exceptionally fast and accurate neural processing; which simultaneously adapts to continuous changes of external inputs and evaluates the flux of top-down information from higher cognitive areas. Performing such almost instantaneous robust pattern recognition is a very challenging task, not only for neurocomputational models but also for the state-of-the-art time series prediction algorithms.

The nature of this incredibly precise machinery is still poorly understood. However, efforts in time series analyses of non-stationary neural recordings during the last decade could provide relevant insights for designing a new class of highly adaptive predictive models for rapidly varying underlying probability distributions. Therefore, neural dynamics analyses and models may enable us to build algorithms which perform with high precision and generalization capability in real-life, non-stationary environments.

In this special session we propose the contribution of works which are focused in temporal pattern recognition on particularly challenging situations; which aim to go beyond out-of-sample performance measures on relatively stationary test datasets.

We welcome contributions about models which are capable of a very high-performance in recognizing and identifying temporal dynamics on non-stationary datasets; such as neurobiological systems can do. Those algorithms should be particularly valuable for accurate time series prediction even when statistical properties of variables change significantly; as it happens for instance in certain industrial settings.

 

Topics of interest include, but are not limited to:

 

• Neurocognitive Dynamics
• Perceptual Processing and Predictive Neural Coding
• Nonlinear Time Series Analysis and Prediction
• Computational Models for Time Series Forecasting:
• Neural Networks (Artificial Neural Networks of varied architectures, Neural Ensemble Models, Spiking Neural Networks, Biophysical Models)
• Computational Statistical Models & Machine Learning (Generative Bayesian Models, Kernel Machines)
• Evolutionary Algorithms
• Fuzzy Modelling

o   Chaos Theory Approaches, etc.

• Change and Novelty Detection in Non-stationary Environments
• Inspiring and novel applications in various domains such as:
• Finance
• Targeted Marketing
• Industry
• Web Mining
• Multivariate Neural Recordings (e.g. Multiple-Unit Recordings, fMRI, EGG, MEG, etc.)
• Traffic
• Sensor Networks
• Smart Environments, etc.

 

Organizers:

Emili Balaguer Ballester^{1, 2} (eb-ballester@bournemouth.ac.uk)

Abdelhamid Bouchachia¹ (abouchachia@bournemouth.ac.uk)

¹Smart Technology Research Center,

School of Design, Engineering and Computing, Bournemouth University, UK.

²Bernstein Center for Computational Neuroscience Heidelberg-Mannheim, Germany.

 

This email is intended only for the person to whom it is addressed and may contain confidential information. If you have received this email in error, please notify the sender and delete this email, which must not be copied, distributed or disclosed to any other person.

Any views or opinions presented are solely those of the author and do not necessarily represent those of Bournemouth University or its subsidiary companies. Nor can any contract be formed on behalf of the University or its subsidiary companies via email.