I am a researcher at Enginyeria i Arquitectura La Salle - Universitat Ramon Llull who is currently finishing his PhD in such an interesting field as the Boltzmann Machine neural network.
At this moment, I am working in a Government funded project which is known as GAD (Active Load Side Managemnent, as standing for its Spanish acornym, Gestión Activa de la Demanda) project, under the National Strategic Consortium for Technical Research (R&D CENIT) initiative. The aim of the project is to acquire knowledge about electric grid optimisation objectives such as peak reduction (demand shifting), energy efficiency, demand response and environmental impact reduction; the project is also expected to establish the different tariff policies, communication protocols and to design the electrical devices that might be used for a proper project deployment. In order to achieve such results, GAD pools the efforts of 15 Spanish companies and 14 research centres into a single stable consortium, which is led by Iberdrola.
However, my spare time is entirely devoted to the Boltzmann Machine (BM) topic. A Boltzmann Machine is a neural network with the ability of learning and extrapolating probability distributions. The original model was born as a parallel implementation of the Simulated Annealing optimization algorithm, but it was later shown that it could be applied a leanring algorithm to become a Hopfield like model. Learning in BMs is often carried out by a gradient descent process that needs some Monte Carlo simulations over the neural network to compute the quantitites needed to update its connections; this makes the learning process slow. We are currently working in some mathematical methods that can be used to speed up this process for any BM topology.