From computer engineering and computer science to artificial intelligence

Early June saw the celebration of the Interdisciplinary Summer School on Artificial Intelligence, with talks on diverse subjects that went from creativity in Twitter bots by Tony Veale to my own talk on computer science and engineering and how they can help AI.

The talk had two parts; the first focused on how AI is improving in performance and decreasing its energy footprint via design of specific chips, many of them based on the RISC-V open hardware architecture. The second part was mainly devoted to concurrent programming and how algorithms and whole applications must be changed to meet the challenges of creating cloud-native programs.

Our research group fully supports free software and open science, and the field of AI is ratcheting up its achievements by working on free stacks, from free hardware to the whole set of programs and services that support them. Understanding these stacks will help us design better algorithms and frameworks in the future.

Anuncios

Apariciones en prensa del Proyecto Sipesca

Recientemente hemos inaugurado en la web de Sipesca [1] una sección con las apariciones en prensa del proyecto, para recoger la difusión que está teniendo el proyecto en los medios.

[1] http://sipesca.ugr.es/prensa/

 

Imagen

Imagen

Las GPUs no son para jugar

O al menos no todo el tiempo, también podemos hacer cosas ‘de provecho’ con ellas… :D

Como expuse en la reunión del pasado Viernes 24, ultimamente se estan adaptando y desarrollando muchos de los algoritmos que todos conocemos (AGs, GPs,  RNs, PSOs, etc), para el aprovechamiento de estas unidades de procesamiento, optimizadas para la computación en paralelo (de forma masiva, ya que cuentan con decenas o cientos de procesadores dedicados al cálculo en coma flotante).

En la siguiente presentación, se exponen sus ventajas e inconvenientes, así como se muestran algunas de las herramientas y APIs utilizadas para realizar dichas implementaciones, como por ejemplo CUDA.

Que la disfruteis! ;) :D

Proyectos fin de carrera 2010-2011

Se acerca el momento de ir buscando tutor de proyecto fin de carrera. Quien quiera hacerlo conmigo o algún otro profesor del grupo GeNeura, y siempre que esté dispuesto a liberar el código resultante (y, si quiere, a participar en el Concurso Universitario de Software Libre, lo puede hacer en una de las siguientes áreas

  • Algoritmos evolutivos aplicados a juegos: resolución del juego del Mastermind, programación de bots en Unreal Tournament, o creación de niveles y de IA para Super Mario
  • Optimización de visualización de resultados de búsquedas, a partir del proyecto Visuse
  • Computación parásita: ejecución de programas masivamente paralelos en el navegador
  • Implementación de algoritmos evolutivos: adaptación y ampliación de la librería Algorithm::Evolutionary (en Perl)
  • Análisis de redes complejas: redes de pases del fútbol, redes de coautorías, otras redes

Si alguien está interesado en convertir alguna de estas ideas en su proyecto fin de carrera, que contacte conmigo.

Thesis on Peer-to-Peer Evolutionary Computation

Last 27th May, I had the dissertation of my thesis entitled: “Peer-to-Peer Evolutionary Computation: A Study of Viability“. It analyzes the viability of the Peer-to-Peer Evolutionary Computation concept and uses to that aim a P2P system as a substrate platform for a parallel implementation of a spatially-structured EA. Dynamics of the P2P platform are extensively described as well as their interactions with the parallel EA which demonstrates good scalability and resilience to the degradation of the system.

Hope that you enjoy the read (if you are crazy enough to go ahead) as I did in the write.

[IJHPSA] Resilience to Churn of a Peer-to-Peer Evolutionary Algorithm

In this paper we analyse the resilience of a Peer-to-Peer (P2P) Evolutionary Algorithm (EA) subject to the following dynamics: computing nodes acting as peers leave the system independently from each other causing a collective effect known as churn. Since the P2P EA has been designed to tackle large instances of computationally expensive problems, we will assess its behaviour under these conditions, by performing a scalability analysis in five different scenarios using the Massively Multimodal Deceptive Problem as a benchmark.In all cases, the P2P EA reaches the success criterion without a penalty on the runtime. We show that the key to the algorithm resilience is to ensure enough peers at the beginning of the experiment; even if some of them leave, those that remain contain enough information to guarantee a reliable convergence.


[PACT’08][PABA Workshop I] Addressing Churn in P2P EA

This week the first Workshop on Parallel Architerctures and Bioinspired Algorithms is being held in Toronto (Canada) in conjunction with the prestigious conference Parallel Architectures and Compilation Techniques (PACT).

In extension to our line of work in P2P EAs, we have presented the work:

In this paper we analyse the robustness of a Peer-to-Peer (P2P) Evolutionary Algorithm (EA) subject to the following dynamics: peers leave the system independently from each other causing a collective effect known as churn. The algorithm has been designed to tackle large instances of computationally expensive problems and, in this paper, we will assess its behavior under churn. To that end, we have performed a scalability analysis in five different scenarios using the Massively Multimodal Deceptive Problem as a
benchmark.  In all cases, the P2P EA reaches the success criterion without a penalty on the response time. The key to the algorithm robustness is to ensure enough peers at the beginning of the experiment. Some of them leave but those that remain are enough to guarantee a reliable  convergence.