Enhancing weak signal transmission through a feedforward network

The ability to transmit and amplify weak signals is fundamental to signal processing of artificial devices in engineering. Using a multilayer feedforward network of coupled double-well oscillators as well as Fitzhugh-Nagumo oscillators, we here investigate the conditions under which a weak signal received by the first layer can be transmitted through the network with or without amplitude attenuation. We find that the coupling strength and the nodes' states of the first layer act as two-state switches, which determine whether the transmission is significantly enhanced or exponentially decreased. We hope this finding is useful for designing artificial signal amplifiers.

Network-based high level data classification

Traditional supervised data classification considers only physical features (e. g., distance or similarity) of the input data. Here, this type of learning is called low level classification. On the other hand, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is, here, referred to as high level classification. In this paper, we propose a hybrid classification technique that combines both types of learning. The low level term can be implemented by any classification technique, while the high level term is realized by the extraction of features of the underlying network constructed from the input data. Thus, the former classifies the test instances by their physical features or class topologies, while the latter measures the compliance of the test instances to the pattern formation of the data. Our study shows that the proposed technique not only can realize classification according to the pattern formation, but also is able to improve the performance of traditional classification techniques. Furthermore, as the class configuration's complexity increases, such as the mixture among different classes, a larger portion of the high level term is required to get correct classification. This feature confirms that the high level classification has a special importance in complex situations of classification. Finally, we show how the proposed technique can be employed in a real-world application, where it is capable of identifying variations and distortions of handwritten digit images. As a result, it supplies an improvement in the overall pattern recognition rate.

Stochastic competitive learning in complex networks

Competitive learning is an important machine learning approach which is widely employed in artificial neural networks. In this paper, we present a rigorous definition of a new type of competitive learning scheme realized on large-scale networks. The model consists of several particles walking within the network and competing with each other to occupy as many nodes as possible, while attempting to reject intruder particles. The particle's walking rule is composed of a stochastic combination of random and preferential movements. The model has been applied to solve community detection and data clustering problems. Computer simulations reveal that the proposed technique presents high precision of community and cluster detections, as well as low computational complexity. Moreover, we have developed an efficient method for estimating the most likely number of clusters by using an evaluator index that monitors the information generated by the competition process itself. We hope this paper will provide an alternative way to the study of competitive learning.

Structure learning of graphical models for task-oriented robot grasping

In the collective imaginaries a robot is a human like machine as any androids in science fiction. However the type of robots that you will encounter most frequently are machinery that do work that is too dangerous, boring or onerous. Most of the robots in the world are of this type. They can be found in auto, medical, manufacturing and space industries. Therefore a robot is a system that contains sensors, control systems, manipulators, power supplies and software all working together to perform a task. The development and use of such a system is an active area of research and one of the main problems is the development of interaction skills with the surrounding environment, which include the ability to grasp objects. To perform this task the robot needs to sense the environment and acquire the object informations, physical attributes that may influence a grasp. Humans can solve this grasping problem easily due to their past experiences, that is why many researchers are approaching it from a machine learning perspective finding grasp of an object using information of already known objects. But humans can select the best grasp amongst a vast repertoire not only considering the physical attributes of the object to grasp but even to obtain a certain effect. This is why in our case the study in the area of robot manipulation is focused on grasping and integrating symbolic tasks with data gained through sensors. The learning model is based on Bayesian Network to encode the statistical dependencies between the data collected by the sensors and the symbolic task. This data representation has several advantages. It allows to take into account the uncertainty of the real world, allowing to deal with sensor noise, encodes notion of causality and provides an unified network for learning. Since the network is actually implemented and based on the human expert knowledge, it is very interesting to implement an automated method to learn the structure as in the future more tasks and object features can be introduced and a complex network design based only on human expert knowledge can become unreliable. Since structure learning algorithms presents some weaknesses, the goal of this thesis is to analyze real data used in the network modeled by the human expert, implement a feasible structure learning approach and compare the results with the network designed by the expert in order to possibly enhance it.

Machine learning methods for prediction of disulphide bonding states of cysteine residues in proteins

The goal of this thesis work is to develop a computational method based on machine learning techniques for predicting disulfide-bonding states of cysteine residues in proteins, which is a sub-problem of a bigger and yet unsolved problem of protein structure prediction. Improvement in the prediction of disulfide bonding states of cysteine residues will help in putting a constraint in the three dimensional (3D) space of the respective protein structure, and thus will eventually help in the prediction of 3D structure of proteins. Results of this work will have direct implications in site-directed mutational studies of proteins, proteins engineering and the problem of protein folding. We have used a combination of Artificial Neural Network (ANN) and Hidden Markov Model (HMM), the so-called Hidden Neural Network (HNN) as a machine learning technique to develop our prediction method. By using different global and local features of proteins (specifically profiles, parity of cysteine residues, average cysteine conservation, correlated mutation, sub-cellular localization, and signal peptide) as inputs and considering Eukaryotes and Prokaryotes separately we have reached to a remarkable accuracy of 94% on cysteine basis for both Eukaryotic and Prokaryotic datasets, and an accuracy of 90% and 93% on protein basis for Eukaryotic dataset and Prokaryotic dataset respectively. These accuracies are best so far ever reached by any existing prediction methods, and thus our prediction method has outperformed all the previously developed approaches and therefore is more reliable. Most interesting part of this thesis work is the differences in the prediction performances of Eukaryotes and Prokaryotes at the basic level of input coding when ‘profile’ information was given as input to our prediction method. And one of the reasons for this we discover is the difference in the amino acid composition of the local environment of bonded and free cysteine residues in Eukaryotes and Prokaryotes. Eukaryotic bonded cysteine examples have a ‘symmetric-cysteine-rich’ environment, where as Prokaryotic bonded examples lack it.

Machine-learning methods for structure prediction of β-barrel membrane proteins

Different types of proteins exist with diverse functions that are essential for living organisms. An important class of proteins is represented by transmembrane proteins which are specifically designed to be inserted into biological membranes and devised to perform very important functions in the cell such as cell communication and active transport across the membrane. Transmembrane β-barrels (TMBBs) are a sub-class of membrane proteins largely under-represented in structure databases because of the extreme difficulty in experimental structure determination. For this reason, computational tools that are able to predict the structure of TMBBs are needed. In this thesis, two computational problems related to TMBBs were addressed: the detection of TMBBs in large datasets of proteins and the prediction of the topology of TMBB proteins. Firstly, a method for TMBB detection was presented based on a novel neural network framework for variable-length sequence classification. The proposed approach was validated on a non-redundant dataset of proteins. Furthermore, we carried-out genome-wide detection using the entire Escherichia coli proteome. In both experiments, the method significantly outperformed other existing state-of-the-art approaches, reaching very high PPV (92%) and MCC (0.82). Secondly, a method was also introduced for TMBB topology prediction. The proposed approach is based on grammatical modelling and probabilistic discriminative models for sequence data labeling. The method was evaluated using a newly generated dataset of 38 TMBB proteins obtained from high-resolution data in the PDB. Results have shown that the model is able to correctly predict topologies of 25 out of 38 protein chains in the dataset. When tested on previously released datasets, the performances of the proposed approach were measured as comparable or superior to the current state-of-the-art of TMBB topology prediction.

LPS-induced modifications in spontaneous network activity causes neuronal apoptosis in neonatal cerebral cortex

During the perinatal period the developing brain is most vulnerable to inflammation. Prenatal infection or exposure to inflammatory factors can have a profound impact on fetal neurodevelopment with long-term neurological deficits, such as cognitive impairment, learning deficits, perinatal brain damage and cerebral palsy. Inflammation in the brain is characterized by activation of resident immune cells, especially microglia and astrocytes whose activation is associated with a variety of neurodegenerative disorders like Alzheimer´s disease and Multiple sclerosis. These cell types express, release and respond to pro-inflammatory mediators such as cytokines, which are critically involved in the immune response to infection. It has been demonstrated recently that cytokines also directly influence neuronal function. Glial cells are capable of releaseing the pro-inflammatory cytokines MIP-2, which is involved in cell death, and tumor necrosis factor alpha (TNFalpha), which enhances excitatory synaptic function by increasing the surface expression of AMPA receptors. Thus constitutively released TNFalpha homeostatically regulates the balance between neuronal excitation and inhibition in an activity-dependent manner. Since TNFalpha is also involved in neuronal cell death, the interplay between neuronal activity MIP-2 and TNFalpha may control the process of cell death and cell survival in developing neuronal networks. An increasing body of evidence suggests that neuronal activity is important in the regulation of neuronal survival during early development, e.g. programmed cell death (apoptosis) is augmented when neuronal activity is blocked. In our study we were interested on the impact of inflammation on neuronal activity and cell survival during early cortical development. To address this question, we investigated the impact of inflammation on neuronal activity and cell survival during early cortical development in vivo and in vitro. Inflammation was experimentally induced by application of the endotoxin lipopolysaccharide (LPS), which initiates a rapid and well-characterized immune response. I studied the consequences of inflammation on spontaneous neuronal network activity and cell death by combining electrophysiological recordings with multi-electrode arrays and quantitative analyses of apoptosis. In addition, I used a cytokine array and antibodies directed against specific cytokines allowing the identification of the pro-inflammatory factors, which are critically involved in these processes. In this study I demonstrated a direct link between inflammation-induced modifications in neuronal network activity and the control of cell survival in a developing neuronal network for the first time. Our in vivo and in vitro recordings showed a fast LPS-induced reduction in occurrence of spontaneous oscillatory activity. It is indicated that LPS-induced inflammation causes fast release of proinflammatory factors which modify neuronal network activity. My experiments with specific antibodies demonstrate that TNFalpha and to a lesser extent MIP-2 seem to be the key mediators causing activity-dependent neuronal cell death in developing brain. These data may be of important clinical relevance, since spontaneous synchronized activity is also a hallmark of the developing human brain and inflammation-induced alterations in this early network activity may have a critical impact on the survival of immature neurons.

Segmentazione e categorizzazione di oggetti mediante immagini depth e deep learning

Questo lavoro è iniziato con uno studio teorico delle principali tecniche di classificazione di immagini note in letteratura, con particolare attenzione ai più diffusi modelli di rappresentazione dell’immagine, quali il modello Bag of Visual Words, e ai principali strumenti di Apprendimento Automatico (Machine Learning). In seguito si è focalizzata l’attenzione sulla analisi di ciò che costituisce lo stato dell’arte per la classificazione delle immagini, ovvero il Deep Learning. Per sperimentare i vantaggi dell’insieme di metodologie di Image Classification, si è fatto uso di Torch7, un framework di calcolo numerico, utilizzabile mediante il linguaggio di scripting Lua, open source, con ampio supporto alle metodologie allo stato dell’arte di Deep Learning. Tramite Torch7 è stata implementata la vera e propria classificazione di immagini poiché questo framework, grazie anche al lavoro di analisi portato avanti da alcuni miei colleghi in precedenza, è risultato essere molto efficace nel categorizzare oggetti in immagini. Le immagini su cui si sono basati i test sperimentali, appartengono a un dataset creato ad hoc per il sistema di visione 3D con la finalità di sperimentare il sistema per individui ipovedenti e non vedenti; in esso sono presenti alcuni tra i principali ostacoli che un ipovedente può incontrare nella propria quotidianità. In particolare il dataset si compone di potenziali ostacoli relativi a una ipotetica situazione di utilizzo all’aperto. Dopo avere stabilito dunque che Torch7 fosse il supporto da usare per la classificazione, l’attenzione si è concentrata sulla possibilità di sfruttare la Visione Stereo per aumentare l’accuratezza della classificazione stessa. Infatti, le immagini appartenenti al dataset sopra citato sono state acquisite mediante una Stereo Camera con elaborazione su FPGA sviluppata dal gruppo di ricerca presso il quale è stato svolto questo lavoro. Ciò ha permesso di utilizzare informazioni di tipo 3D, quali il livello di depth (profondità) di ogni oggetto appartenente all’immagine, per segmentare, attraverso un algoritmo realizzato in C++, gli oggetti di interesse, escludendo il resto della scena. L’ultima fase del lavoro è stata quella di testare Torch7 sul dataset di immagini, preventivamente segmentate attraverso l’algoritmo di segmentazione appena delineato, al fine di eseguire il riconoscimento della tipologia di ostacolo individuato dal sistema.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Reinforcement learning in dendritic structures

The discovery of binary dendritic events such as local NMDA spikes in dendritic subbranches led to the suggestion that dendritic trees could be computationally equivalent to a 2-layer network of point neurons, with a single output unit represented by the soma, and input units represented by the dendritic branches. Although this interpretation endows a neuron with a high computational power, it is functionally not clear why nature would have preferred the dendritic solution with a single but complex neuron, as opposed to the network solution with many but simple units. We show that the dendritic solution has a distinguished advantage over the network solution when considering different learning tasks. Its key property is that the dendritic branches receive an immediate feedback from the somatic output spike, while in the corresponding network architecture the feedback would require additional backpropagating connections to the input units. Assuming a reinforcement learning scenario we formally derive a learning rule for the synaptic contacts on the individual dendritic trees which depends on the presynaptic activity, the local NMDA spikes, the somatic action potential, and a delayed reinforcement signal. We test the model for two scenarios: the learning of binary classifications and of precise spike timings. We show that the immediate feedback represented by the backpropagating action potential supplies the individual dendritic branches with enough information to efficiently adapt their synapses and to speed up the learning process.

Reinforcement learning in dendritic structures

The discovery of binary dendritic events such as local NMDA spikes in dendritic subbranches led to the suggestion that dendritic trees could be computationally equivalent to a 2-layer network of point neurons, with a single output unit represented by the soma, and input units represented by the dendritic branches. Although this interpretation endows a neuron with a high computational power, it is functionally not clear why nature would have preferred the dendritic solution with a single but complex neuron, as opposed to the network solution with many but simple units. We show that the dendritic solution has a distinguished advantage over the network solution when considering different learning tasks. Its key property is that the dendritic branches receive an immediate feedback from the somatic output spike, while in the corresponding network architecture the feedback would require additional backpropagating connections to the input units. Assuming a reinforcement learning scenario we formally derive a learning rule for the synaptic contacts on the individual dendritic trees which depends on the presynaptic activity, the local NMDA spikes, the somatic action potential, and a delayed reinforcement signal. We test the model for two scenarios: the learning of binary classifications and of precise spike timings. We show that the immediate feedback represented by the backpropagating action potential supplies the individual dendritic branches with enough information to efficiently adapt their synapses and to speed up the learning process.

Exploring the effectiveness of development research with a monitoring and learning approach

There is a growing demand for better understanding of the link between research, policy and practice in development. This article provides findings from a study that aimed to gain insights into how researchers engage with their non-academic partners. It draws on experiences from the National Centre of Competence in Research North-South programme, a development research network of Swiss, African, Asian and Latin American institutions. Conceptually, this study is concerned with research effectiveness as a means to identify knowledge useful for society. Research can be improved and adapted when monitoring the effects of interactions between researchers and non-academic partners. Therefore, a monitoring and learning approach was chosen. This study reveals researchers' strategies in engaging with non-academic partners and points to framing conditions considered decisive for soccessful interactions. It concludes that reserachrs need to systematically analyse the socio-political context in which they intervene. By providing insights from the ground and reflecting on them in the light of the latest theoretical concepts, this article contributes to the emerging literature founded on practice-based experience.

Functional neuroimaging of emotional learning and autonomic reactions

This article provides a selective overview of the functional neuroimaging literature with an emphasis on emotional activation processes. Emotions are fast and flexible response systems that provide basic tendencies for adaptive action. From the range of involved component functions, we first discuss selected automatic mechanisms that control basic adaptational changes. Second, we illustrate how neuroimaging work has contributed to the mapping of the network components associated with basic emotion families (fear, anger, disgust, happiness), and secondary dimensional concepts that organise the meaning space for subjective experience and verbal labels (emotional valence, activity/intensity, approach/withdrawal, etc.). Third, results and methodological difficulties are discussed in view of own neuroimaging experiments that investigated the component functions involved in emotional learning. The amygdala, prefrontal cortex, and striatum form a network of reciprocal connections that show topographically distinct patterns of activity as a correlate of up and down regulation processes during an emotional episode. Emotional modulations of other brain systems have attracted recent research interests. Emotional neuroimaging calls for more representative designs that highlight the modulatory influences of regulation strategies and socio-cultural factors responsible for inhibitory control and extinction. We conclude by emphasising the relevance of the temporal process dynamics of emotional activations that may provide improved prediction of individual differences in emotionality.

Hippocampal functional connectivity reflects verbal episodic memory network integrity

Using functional magnetic resonance imaging during a verbal memory task, we investigated correlations of signal fluctuations within the hippocampus and ipsilateral frontal as well as temporal areas in temporal lobe epilepsy patients. Declarative memory abilities were additionally examined before and after temporal lobe epilepsy surgery. A significant difference exists in functional connectivity between patients whose mnemonic functions deteriorated and those who remained stable or improved. Univariate analyses showed significantly higher preoperative coupling between the hippocampus and Brodmann area 22 for the group that decreased in verbal learning. We suggest greater coupling to reflect higher functional network integrity. Postoperatively reduced learning ability in patients with higher preoperative coupling underlines the importance of hippocampal interaction with cortical areas for successful memory formation.

IDENTIFICATION OF GENES CONTROLLING BIOLOGICAL PROCESSES AND PATHWAYS THROUGH STATISTICAL ANALYSIS AND NETWORK RECONSTRUCTION

The developmental processes and functions of an organism are controlled by the genes and the proteins that are derived from these genes. The identification of key genes and the reconstruction of gene networks can provide a model to help us understand the regulatory mechanisms for the initiation and progression of biological processes or functional abnormalities (e.g. diseases) in living organisms. In this dissertation, I have developed statistical methods to identify the genes and transcription factors (TFs) involved in biological processes, constructed their regulatory networks, and also evaluated some existing association methods to find robust methods for coexpression analyses. Two kinds of data sets were used for this work: genotype data and gene expression microarray data. On the basis of these data sets, this dissertation has two major parts, together forming six chapters. The first part deals with developing association methods for rare variants using genotype data (chapter 4 and 5). The second part deals with developing and/or evaluating statistical methods to identify genes and TFs involved in biological processes, and construction of their regulatory networks using gene expression data (chapter 2, 3, and 6). For the first part, I have developed two methods to find the groupwise association of rare variants with given diseases or traits. The first method is based on kernel machine learning and can be applied to both quantitative as well as qualitative traits. Simulation results showed that the proposed method has improved power over the existing weighted sum method (WS) in most settings. The second method uses multiple phenotypes to select a few top significant genes. It then finds the association of each gene with each phenotype while controlling the population stratification by adjusting the data for ancestry using principal components. This method was applied to GAW 17 data and was able to find several disease risk genes. For the second part, I have worked on three problems. First problem involved evaluation of eight gene association methods. A very comprehensive comparison of these methods with further analysis clearly demonstrates the distinct and common performance of these eight gene association methods. For the second problem, an algorithm named the bottom-up graphical Gaussian model was developed to identify the TFs that regulate pathway genes and reconstruct their hierarchical regulatory networks. This algorithm has produced very significant results and it is the first report to produce such hierarchical networks for these pathways. The third problem dealt with developing another algorithm called the top-down graphical Gaussian model that identifies the network governed by a specific TF. The network produced by the algorithm is proven to be of very high accuracy.