Attendance at the International Joint Conference on Neural Networks 2013

Tomorrow I’ll be attending the International joint Conference on Neural Networks. This is one of the most important events in the area of Artificial Intelligence, and will be performed in the City of Dallas, TX.

A huge variety of  tutorials will be presented such as “Cognitive Computational Intelligence”, “Random Neural Network and Applications in Engineering and Biology”, and many more. Also, I’ll be attending the workshops: “What Language And Emotion Can Tell Us About the Brain: New Methods of Analysis” and “Perception and Cognition in the Brain: Integrating Single Cell Recordings, Spiking Neurons, and a Brain Theory”. Which are both quite interesting for my research in neural models.

In this event, I’ll present a new training method for Polynomial Cellular Neural Networks (PCNN): These models are capable of implementing powerful classification mechanisms, Implementations of cellular automata, and even the hypothesis that these models are capable of computational universality in the Turing Sense. In other words, If a polynomial cellular neural network is capable of achieving the computational universality, it is able to execute any algorithm, and be equivalent to any general purpose computer.

The first step toward this goal is to implement any cellular automata in a PCNN, using a single layer of neurons. But, the main challenge is the determination of the synaptic weights: If you are able to choose appropriately the synaptic weights of a PCNN, you can achieve any behavior you desire.

In the presentation, a root location training method will be described: The algorithm receives the behavior as an input and straightforwardly obtains the synaptic weights that implements your desired behavior. For now, the implementations are limited to the Totalistic Cellular Automata behaviors, but future work will extend this limit to the Semitotalistic and Universal Cellular Automata behaviors.

The paper (and proof of the theorem that was mentioned in the conference) can be found here: A root location training method for PCNN

After the presentation, the slides will be available here. (and perhaps a video ;D ).

Here is a brief list of another cool things you can model and build with Cellular Automata and Cellular Neural Networks (which are not my work, all credits goes to their respective authors) :


Attendance and conference in the International CCE 2012 (Mexico City)

Tomorrow I will be attending the 9th International CCE conference at Mexico City. My topic will verse about the usage of artificial neural networks as powerful auxiliary  tools in the diagnosis of malnutrition related diseases.

This paper was written with the aid of my father, who is a great Doctor, Pediatrician and quite a visionary. His work is vast and amusing in several areas of medicine such as Antibiotics, Nutrition, Infectology, Immunology and general Pediatrics. Therefore the medical background of this work was quite solid. (And in the research it was quite notorious! His experience and knowledge were molded perfectly in the resolution of my doubts.)

Why this topic? Why do we have to care about the malnutrition related diseases? We have to take into account that a child that manifests chronic malnutrition may be handicapped both intellectually and physically in the future. A child that suffers from malnutrition may be handicapped in the very early stages of his life.

Sometimes the malnutrition is related to feeding disorders (infantile anorexia, fear of feeding, etc.) and it has been proposed several methods to aid the pediatrician in his labor in order to eradicate these disorders(I worked in a project for Abbott Laboratories in this particual topic).

Nonetheless if the root of the problem is an organic disease, it may be that the symptomatology is quite obvious. But if it is not, the patient may suffer malnutrition after malnutrition, and if the disease is not cured or at least identified, the patient will fail to thrive in his/her development.

Therefore, we proposed a conjunction of Artificial Neural Networks / Self Organizing Map mechanisms in order to aid the  pediatrician in his clinical labor, helping not only to establish a diagnosis, but also do it in a quick, accurate and reliable manner.

The work will be published in the “Artificial Neural Networks” section. Enjoy!

Antonio Arista Jalife

The International Joint Conference on Neural Networks 2012

Last week I had the honor to be part of the International Joint Conference on Neural Networks (IJCNN) at the international conference centre in Brisbane, Australia. Also, the FUZZ and CEC events were merged in the event. These set of conferences is called World Congress of Computational Intelligence (WCCI). And it is organized every two years.

For me it was an outstanding experience, it was a good opportunity to gather knowledge and experiences from all over the world in terms of computational intelligence, the conferences about evolutionary computing and spiking neural networks were quite interesting and inspiring in the creation and application of new ideas and resources.

I was quite amused by a fact that Nikola Kasabov, one of the plenary speakers, explained in his exposition: A single spiking neuron, with the proper training, is able to recognize 20, or even 30 different patterns, with an acceptable success rate. Not only the fact its interesting, the reason it amused me so much is that, a colleague and friend of mine (Aleister Cachón) was able to achieve the same conclusion some time before in his thesis. 

Here are some photos of the event:


My presentation:



Fast simulation of spiking neurons.

We have managed to create fast simulators of spiking neural networks, which are capable of modelling biologically realistic processes. In this case we have managed to create 50,000 neurons, with 1,000 synapses each one, the simulation was completed in 12.27 seconds using NVIDIA’s Common Unified Device Architecture.

In this video we are representing excitatory neurons as green dots and inhibitory neurons as blue dots. each time a dot is shown on the screen it means that a neuron has reached a spiking status. As you can notice it presents alpha / gamma rythms, just as the Izhikevich’s model.

I’ll be presenting this simulation at the International Joint Conference of Neural Networks (IJCNN) 2012, at Brisbane, Australia.

You can check more information about this research here.


A simple face localization program.

Face localization and recognition is usually found in cameras and security devices, in this post I’ll show the results of a simple face recognition program that can pinpoint faces and skin sections and, with further programming, it may be able to recognize faces.

Write me a mail at if you are interested in the source code 😀

Some pictures of its performance:

Picture 1: The author and the recognition 😉


Picture 2: Some friends.


Picture 3: Other friends



Welcome to “Cybernetics as an art”! As the subtitle says, in this blog I’ll post some of my ideas, concepts, notes and thoughts. There are a couple of reasons to do so: Feedback from all over the world, collaboration opportunities may rise, even good ideas may born between lines and comments.

Why consider cybernetics as an art? Art is defined in the Encyclopedia Britannica as:

The use of skill and imagination in the creation of aesthetic objects, environments, or experiences that can be shared with others.

I firmly believe that cybernetics and computer systems are able to achieve this role too, they can be functional and at the same time recreate experiences that can be freely shared. In a world where science, math and technology seems divorced from the aesthetic branches of humanity, perhaps it can be a good moment to pose a bridge that connects these branches.

I hope that you enjoy reading this blog as much as I enjoy writing it!