El acceso a la versión del editor puede requerir la suscripción del recurso Access to the published version may require subscription AbstractThe learning process in Boltzmann Machines is computationally very expensive. The computational complexity of the exact algorithm is exponential in the number of neurons. We present a new approximate learning algorithm for Boltzmann Machines, which is based on mean eld theory and the linear response theorem. The computational complexity of the algorithm is cubic in the number of neurons.In the absence of hidden units, we show how the weights can be directly computed from the xed point equation of the learning rules. Thus, in this case we do not need to use a gradient descent procedure for the learning process. We show that the solutions of this method are close to the optimal solutions and give a signi cant improvement when correlations play a signi cant role. Finally, we apply the method to a pattern completion task and show good performance for networks up to 100 neurons.

Recent experiments have revealed the existence of neural signatures in the activity of individual cells of the pyloric central pattern generator (CPG) of crustacean. The neural signatures consist of cell-specific spike timings in the bursting activity of the neurons. The role of these intraburst neural fingerprints is still unclear. It has been reported previously that some muscles can reflect small changes in the spike timings of the neurons that innervate them. However, it is unclear to what extent neural signatures contribute to the command message that the muscles receive from the motoneurons. It is also unknown whether the signatures have any functional meaning for the neurons that belong to the same CPG or to other interconnected CPGs. In this paper, we use realistic neural models to study the ability of single cells and small circuits to recognize individual neural signatures. We show that model cells and circuits can respond distinctly to the incoming neural fingerprints in addition to the properties of the slow depolarizing waves. Our results suggest that neural signatures can be a general mechanism of spiking-bursting cells to implement multicoding.

The idea of closed-loop interaction in in vitro and in vivo electrophysiology has been successfully implemented in the dynamic clamp concept strongly impacting the research of membrane and synaptic properties of neurons. In this paper we show that this concept can be easily generalized to build other kinds of closed-loop protocols beyond (or in addition to) electrical stimulation and recording in neurophysiology and behavioral studies for neuroethology. In particular, we illustrate three different examples of goal-driven real-time closed-loop interactions with drug microinjectors, mechanical devices and video event driven stimulation. Modern activity-dependent stimulation protocols can be used to reveal dynamics (otherwise hidden under traditional stimulation techniques), achieve control of natural and pathological states, induce learning, bridge between disparate levels of analysis and for a further automation of experiments. We argue that closed-loop interaction calls for novel real time analysis, prediction and control tools and a new perspective for designing stimulus-response experiments, which can have a large impact in neuroscience research.

No abstract

A technique that combines a theoretical description of the electrostatic interaction and artificial neural networks (ANNs) is used to solve an inverse problem in scanning probe microscopy setups. Electrostatic interaction curves calculated by the generalized image charge method are used to train and validate the ANN in order to estimate unknown magnitudes in highly undetermined setups. To illustrate this technique, we simultaneously estimate the tip-sample distance and the dielectric constant for a system composed of a tip scanning over a metallic nanowire. In a second example, we use this method to quantitatively estimate the dielectric constant for an even more undetermined system where the tip shape (characterized by three free parameters) is not known. Finally, the proposed method is validated with experimental data.