Francisco de Borja Rodríguez

Efficient learning in Boltzmann machines using linear response theory

The 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, based on mean-field 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 fixed-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 significant improvement when correlations play a significant role. Finally, we apply the method to a pattern completion task and show good performance for networks up to 100 neurons.

Cryptanalysis of a one round chaos-based Substitution Permutation Network

The interleaving of chaos and cryptography has been the aim of a large set of works since the beginning of the nineties. Many encryption proposals have been introduced to improve conventional cryptography. However, many of possess serious problems according to the basic requirements for the secure exchange of information. In this paper we highlight some of the main problems of chaotic cryptography by means of the analysis of a very recent chaotic cryptosystem based on a one round Substitution Permutation Network. More specifically, we show that it is not possible to avoid the security problems of that encryption architecture just by including a chaotic system as the core of the derived encryption system. Highlights? The security problems of one round Substitution Permutation Networks are discussed. ? The problems of considering unimodal maps as the core of cryptosystems are analysed. ? The cryptanalysis of an specific permutation-only cipher is performed.

Neural signatures: multiple coding in spiking-bursting cells

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.

A two-step neural-network based algorithm for fast image super-resolution

We propose a novel, learning-based algorithm for image super-resolution. First, an optimal distance-based weighted interpolation of the image sequence is performed using a new neural architecture, hybrid of a multi-layer perceptron and a probabilistic neural network, trained on synthetic image data. Secondly, a linear filter is applied with coefficients learned to restore residual interpolation artifacts in addition to low-resolution blurring, providing noticeable improvements over lens-detector Wiener restorations. Our method has been evaluated on real visible and IR sequences with widely different contents, providing significantly better results that a two-step method with high computational requirements. Results were similar or better than those of a maximum-a-posteriori estimator, with a reduction in processing time by a factor of almost 300. This paves the way to high-quality, quasi-real time applications of super-resolution techniques.

Controlling a Smartphone Using Gaze Gestures as the Input Mechanism

The emergence of small handheld devices such as tablets and smartphones, often with touch sensitive surfaces as their only input modality, has spurred a growing interest in the subject of gestures for human–computer interaction (HCI). It has been proven before that eye movements can be consciously controlled by humans to the extent of performing sequences of predefined movement patterns, or “gaze gestures” that can be used for HCI purposes in desktop computers. Gaze gestures can be tracked noninvasively using a video-based eye-tracking system. We propose here that gaze gestures can also be an effective input paradigm to interact with handheld electronic devices. We show through a pilot user study how gaze gestures can be used to interact with a smartphone, how they are easily assimilated by potential users, and how the Needleman-Wunsch algorithm can effectively discriminate intentional gaze gestures from otherwise typical gaze activity performed during standard interaction with a small smartphone screen. Hence, reliable gaze–smartphone interaction is possible with accuracy rates, depending on the modality of gaze gestures being used (with or without dwell), higher than 80 to 90%, negligible false positive rates, and completion speeds lower than 1 to 1.5 s per gesture. These encouraging results and the low-cost eye-tracking equipment used suggest the possibilities of this new HCI modality for the field of interaction with small-screen handheld devices.

Extending the bioinspired hierarchical temporal memory paradigm for sign language recognition

Sign language recognition, SLR, using spatial positions and arrangements of the hands over time is a challenging multi-variable time series recognition problem with several potential applications. Here we explore, for SLR purposes, a hierarchically connected network of nodes based on a Bayesian-like paradigm known as hierarchical temporal memory, HTM, that models neocortical principles of organization and information coding. HTM is a broad paradigm for pattern recognition, control, attention and forward prediction that exploits the hierarchy in time and space existing in the physical world during both learning and inference. In this work we focus on HTM capabilities for pattern recognition. We extend the traditional HTM paradigm with an original top node in order to improve HTMs performance in problems where instances unfold over time. The extended top node stores and compares sequences of spatio-temporally codified inputs to handle the temporal evolution of instances in sign language. Sequence comparison is carried out using the Needleman-Wunsch algorithm for sequence alignment that employs dynamic programming. We compare the performance of the extended HTM with traditional HTMs and machine learning algorithms routinely used in the literature for SLR. The extended HTM improves performance of traditional HTM for SLR, reaching 91% recognition accuracy for a data set of 95 categories of Australian sign language. When sufficient training instances are available, the extended HTM matches or outperforms state of the art methods for SLR such as Hidden Markov Models or Metafeatures T-Classes without the usage of a language model, nor pre-processing of sensor data. The extended HTM employs relatively small feature vectors in comparison to methods in the literature. Our method learns the spatio-temporal data structures and transitions that occur in the data without depending on manually predefined features to be searched for and works well in real time. These results suggest that the extended HTM approach is a valid bioinspired alternative to existing SLR engines and that it can be successfully applied to other machine learning tasks whose input instances also unfold over time.

Superresolution imaging: a survey of current techniques

Imaging plays a key role in many diverse areas of application, such as astronomy, remote sensing, microscopy, and tomography. Owing to imperfections of measuring devices (e.g., optical degradations, limited size of sensors) and instability of the observed scene (e.g., object motion, media turbulence), acquired images can be indistinct, noisy, and may exhibit insuffcient spatial and temporal resolution. In particular, several external effects blur images. Techniques for recovering the original image include blind deconvolution (to remove blur) and superresolution (SR). The stability of these methods depends on having more than one image of the same frame. Differences between images are necessary to provide new information, but they can be almost unperceivable. State-of-the-art SR techniques achieve remarkable results in resolution enhancement by estimating the subpixel shifts between images, but they lack any apparatus for calculating the blurs. In this paper, after introducing a review of current SR techniques we describe two recently developed SR methods by the authors. First, we introduce a variational method that minimizes a regularized energy function with respect to the high resolution image and blurs. In this way we establish a unifying way to simultaneously estimate the blurs and the high resolution image. By estimating blurs we automatically estimate shifts with subpixel accuracy, which is inherent for good SR performance. Second, an innovative learning-based algorithm using a neural architecture for SR is described. Comparative experiments on real data illustrate the robustness and utilization of both methods.

Effect of individual spiking activity on rhythm generation of central pattern generators

Abstract Central pattern generators (CPGs) are highly specialized neural networks often with redundant elements that allow the system to act properly in case of error. CPGs are multifunctional circuits, i.e. the same CPG can produce many different rhythms in response to modulatory or sensory inputs. All these rhythms have to be optimal for motor control and coordination. In this paper, we use a model of the well-known pyloric CPG of crustacean to analyze the importance of redundant connections and individual spiking activity in the generation of its rhythm. In particular, we study the effect of different individual spike distributions on the network behavior.

A hybrid MLP-PNN architecture for fast image superresolution

Image superresolution methods process an input image sequence of a scene to obtain a still image with increased resolution. Classical approaches to this problem involve complex iterative minimization procedures, typically with high computational costs. In this paper is proposed a novel algorithm for super-resolution that enables a substantial decrease in computer load. First, a probabilistic neural network architecture is used to perform a scattered-point interpolation of the image sequence data. The network kernel function is optimally determined for this problem by a multi-layer perceptron trained on synthetic data. Network parameters dependence on sequence noise level is quantitatively analyzed. This super-sampled image is spatially filtered to correct finite pixel size effects, to yield the final high-resolution estimate. Results on a real outdoor sequence are presented, showing the quality of the proposed method.

Reducing the Loss of Information through Annealing Text Distortion

Compression distances have been widely used in knowledge discovery and data mining. They are parameter-free, widely applicable, and very effective in several domains. However, little has been done to interpret their results or to explain their behavior. In this paper, we take a step toward understanding compression distances by performing an experimental evaluation of the impact of several kinds of information distortion on compression-based text clustering. We show how progressively removing words in such a way that the complexity of a document is slowly reduced helps the compression-based text clustering and improves its accuracy. In fact, we show how the nondistorted text clustering can be improved by means of annealing text distortion. The experimental results shown in this paper are consistent using different data sets, and different compression algorithms belonging to the most important compression families: Lempel-Ziv, Statistical and Block-Sorting.