Information capacity

In this work, we propose a new type of activation function for a complex valued neural network (CVNN). This activation function is a special Möbius transformation classified as reflection. It is bounded outside of the unit disk and has partial continuous derivatives but not differentiable since it does not satisfy the Cauchy-Riemann equalities. However, the fixed points set of this function is a circle. Therefore, we employ this activation to a specific complex valued Hopfield neural network (CVHNN) and increase the number of fixed points. Using of this activation function leads us also to guarantee the existence of fixed points of the CVHNN. It is shown that the fixed points are all attractive which indicates that storage capacity of the CVHNN is enlarged.

The stochastic resonance (SR) is a phenomenon of nonlinear systems in which the addition of an intermediate level of noise improves the response of such system. Although SR has been studied in isolated hair cells and in the bullfrog sacculus, the occurrence of this phenomenon in the vestibular system in development is unknown. The purpose of the present study was to explore for the existence of SR via natural mechanical-stimulation in the hair cell-vestibular primary afferent transmission. In vitro experiments were performed on the posterior semicircular canal of the chicken inner ear during development. Our experiments showed that the signal-to-noise ratio of the afferent multiunit activity from E15 to P5 stages of development exhibited the SR phenomenon, which was characterized by an inverted U-like response as a function of the input noise level. The inverted U-like graphs of SR acquired their higher amplitude after the post-hatching stage of development. Blockage of the synaptic transmission with selective antagonists of the NMDA and AMPA/Kainate receptors abolished the SR of the afferent multiunit activity. Furthermore, computer simulations on a model of the hair cell-primary afferent synapse qualitatively reproduced this SR behavior and provided a possible explanation of how and where the SR could occur. These results demonstrate that a particular level of mechanical noise on the semicircular canals can improve the performance of the vestibular system in their peripheral sensory processing even during embryonic stages of development.

Two well known, biologically inspired non-dynamical models of stochastic resonance, the threshold-crossing model and the fluctuating rate model, are analyzed in terms of channel information capacity and dissipation of energy necessary for small-signal transduction. Using analogies to spike propagation in neurons we postulate the average output pulse rate as a measure of dissipation. The dissipation increases monotonically with the input noise. We find that for small dissipation both models give a close to linear dependence of the channel information capacity on dissipation. In both models the channel information capacity, as a function of dissipation, has a maximum at input noise amplitude that is different from that in the standard signal-to-noise ratio versus input noise plot. Though a direct comparison is not straightforward, for small signals the threshold model gives appreciably higher density of information per dissipation than the exponential fluctuating rate model. We show that a formal introduction of cooperativity in the rate fluctuating model permits us to imitate the response function of the threshold model and to enhance performance. This finding may have direct relevance to real neural spike generation where, due to a strong positive feedback, the ion channel currents add up in a synchronized way.

Stochastic resonance (SR) is commonly understood to be the enhancement, by noise, of the response of a system to a weak input signal. The aim of this study was to demonstrate the occurrence of SR in spinal and cortical evoked field potentials (EFPs) elicited by periodic tactile stimuli in the anesthetized cat. The electrodes were positioned in spinal and cortical somatosensory regions in which the largest negative EFPs were detected. The periodic tactile stimuli consisted of local skin displacements on the central pad of the hindpaw. Two series of experiments were performed. First, periodic tactile stimuli and the noisy tactile stimuli were applied with the same indenter. Second, noisy tactile stimuli were applied with an additional indenter placed on the glabrous skin of the third hindpaw digit. This last protocol ensured that the signal and noise were mixed not in the skin but in the somatosensory regions of the CNS. All cats showed distinct SR behavior at the spinal and cortical ...

The aim of this study was to demonstrate the occurrence of stochastic resonance (SR) in spinal and cortical potentials elicited by periodic tactile stimuli in the anaesthetised cat. The periodic tactile stimuli were applied on the central pad of the hindpaw and the noisy tactile stimuli on the glabrous skin of the third hindpaw digit. This protocol allowed that the signal and noise were mixed not in the skin but in the somatosensory regions of the central nervous system. The results show that a particular level of tactile noise can increase the amplitude of the spinal and cortical potentials elicited by periodic tactile stimuli. The topographical distribution of evoked potentials indicates that the effects of noise were spatially restricted. All cats showed distinct SR behavior at the spinal and cortical stages of the sensory encoding. Such SR was abolished in the cortical but not in the spinal recording after the sectioning of the ascending pathways. This suggests that the spinal neurones may also contribute to the SR observed at the cortical level. The present study documents the first evidence that the SR phenomenon occurs in the spinal and cortical somatosensory system itself and not only in the peripheral sensory receptors.

Abstract: We present here a cosmological myth, alternative (but complementary) to “the Universe Story” and “the Epic of Evolution”, highlighting the roles of entropy and dissipative structures in the universe inaugurated by the Big Bang. This myth offers answers to these questions: ...

Similarity based fault tolerant retrieval in neural associative memories (N AM) has not lead to wiedespread applications. A drawback of the efficient Willshaw model for sparse patterns [Ste61, WBLH69], is that the high asymptotic information capacity is of little practical use because of high cross talk noise arising in the retrieval for finite sizes. Here a new bidirectional iterative retrieval method for the Willshaw model is presented, called crosswise bidirectional (CB) retrieval, providing enhanced performance. We discuss its asymptotic capacity limit, analyze the first step, and compare it in experiments with the Willshaw model. Applying the very efficient CB memory model either in information retrieval systems or as a functional model for reciprocal cortico-cortical pathways requires more than robustness against random noise in the input: Our experiments show also the segmentation ability of CB-retrieval with addresses containing the superposition of pattens, provided even at high memory load.

A new a m to the asymptotic analysis of autoassociation properties in recurrent McCulloch-Pitts networks in the range of low activity is proposed. Using information theory, this method examines the sialic S l N C t U T e of stable states imprinted by a Hebbian storing process. In addition to the definition of cfitical pattern capacity usually considered in the analysis of the Hopfield model. we introduce the definition of information capacity which guarantees content adressability and is a stricter upper hound of the information really accessible in an autoassociation process. We calculate these two types of capacities faor two types of local learning rules which are very effective for sparsely coded patterns: the Hebb rule and the clipped Hebb rule. It tums out that for both rules the information capacity is exactly half the pattern capacity.

Similarity based fault tolerant retrieval in neural associative memories (N AM) has not lead to wiedespread applications. A drawback of the efficient Willshaw model for sparse patterns [Ste61, WBLH69], is that the high asymptotic information capacity is of little practical use because of high cross talk noise arising in the retrieval for finite sizes. Here a new bidirectional iterative retrieval method for the Willshaw model is presented, called crosswise bidirectional (CB) retrieval, providing enhanced performance. We discuss its asymptotic capacity limit, analyze the first step, and compare it in experiments with the Willshaw model. Applying the very efficient CB memory model either in information retrieval systems or as a functional model for reciprocal cortico-cortical pathways requires more than robustness against random noise in the input: Our experiments show also the segmentation ability of CB-retrieval with addresses containing the superposition of pattens, provided even at high memory load.

The repertoire of neural activity patterns that a cortical network can produce constrains the network's ability to transfer and process information. Here, we measured activity patterns obtained from multi-site local field potential (LFP) recordings in cortex cultures, urethane anesthetized rats, and awake macaque monkeys. First, we quantified the information capacity of the pattern repertoire of ongoing and stimulusevoked activity using Shannon entropy. Next, we quantified the efficacy of information transmission between stimulus and response using mutual information. By systematically changing the ratio of excitation/inhibition (E/I) in vitro and in a network model, we discovered that both information capacity and information transmission are maximized at a particular intermediate E/I, at which ongoing activity emerges as neuronal avalanches. Next, we used our in vitro and model results to correctly predict in vivo information capacity and interactions between neuronal groups during ongoing activity. Close agreement between our experiments and model suggest that neuronal avalanches and peak information capacity arise due to criticality and are general properties of cortical networks with balanced E/I.

The information content of medical images produced by X-ray beam±scintillator interaction was assessed by a method involving detective quantum eciency (DQE) and information capacity determination. The method was based on emitted light ¯uence and modulation transfer function (MTF) measurements and it was employed to evaluate the performance of (Gd 0X5 La 0X5 ) 2 O 2 S:Tb scintillator, which was used in the form of laboratory prepared scintillator screens with coating weights ranging between 20 and 139 mgacm 2 . Screens were excited to luminescence using 80 and 120 kVp X-ray beams. The 83 mgacm 2 scintillator layer exhibited the highest DQE at zero spatial frequency. Thick scintillator layers displayed higher DQE values in the low frequency range while at high frequencies the DQE of thin scintillators were higher. Additionally, thin scintillator layers showed slow DQE variation with increasing frequency. The information capacity of scintillator layers decreased with increasing coating weight indicating that the information content of images produced by thin scintillating layers is higher and more evenly distributed with respect to spatial frequencies.

Discrete chaotic signals with a high information capacity are developed and studied on the basis of nonlinear chaotic dynamic systems. The structural and fractal complexity of pseudorandom integer and binary sequences is analyzed. A model digital spread-spectrum radio channel is used to show experimentally that the chaotic algorithms developed can be applied efficiently in various radio facilities, such as telecommunications, radar, and navigation systems. A technique is developed for forming a reproducible irregular fractal structure involved in frequency-independent fractal antennas with a given fractal dimension. An integer chaotic algorithm with retardation is employed to generate a pseudorandom sequence.

We directly lower bound the information capacity for channels with i.i.d. deletions and duplications. Our approach differs from previous work in that we focus on the information capacity using ideas from renewal theory, rather than focusing on the transmission capacity by analyzing the error probability of some randomly generated code using a combinatorial argument. Of course, the transmission and information capacities are equal, but our change of perspective allows for a much simpler analysis that gives more general theoretical results. We then apply these results to the binary deletion channel to improve existing lower bounds on its capacity.

We directly lower bound the information capacity for channels with i.i.d. deletions and duplications. Our approach differs from previous work in that we focus on the information capacity using ideas from renewal theory, rather than focusing on the transmission capacity by analyzing the error probability of some randomly generated code using a combinatorial argument. Of course, the transmission and information capacities are equal, but our change of perspective allows for a much simpler analysis that gives more general theoretical results. We then apply these results to the binary deletion channel to improve existing lower bounds on its capacity.

We evaluate the information capacities of a lossy bosonic channel with correlated noise. The model generalizes the one recently discussed in [Phys. Rev. A 77, 052324 (2008)], where memory effects come from the interaction with correlated environments. Environmental correlations are quantified by a multimode squeezing parameter, which vanishes in the memoryless limit. We show that a global encoding/decoding scheme, which involves input entangled states among different channel uses, is always preferable with respect to a local one in the presence of memory. Moreover, in a certain range of the parameters, we provide an analytical expression for the classical capacity of the channel showing that a global encoding/decoding scheme allows to attain it. All the results can be applied to a broad class of bosonic Gaussian channels.

We evaluate the information capacities of a lossy bosonic channel with correlated noise. The model generalizes the one recently discussed in [Phys. Rev. A 77, 052324 (2008)], where memory effects come from the interaction with correlated environments. Environmental correlations are quantified by a multimode squeezing parameter, which vanishes in the memoryless limit. We show that a global encoding/decoding scheme, which involves input entangled states among different channel uses, is always preferable with respect to a local one in the presence of memory. Moreover, in a certain range of the parameters, we provide an analytical expression for the classical capacity of the channel showing that a global encoding/decoding scheme allows to attain it. All the results can be applied to a broad class of bosonic Gaussian channels.

This paper explains the home equity bias and its puzzling evolution in a model where investors face an information constraint and have an initial local informational advantage. After …nancial liberalization, local investors have a magni…ed informational advantage since information processed under autarky remains useful after liberalization. A gradual shift towards foreign assets occurs as the relevance of autarkic information declines over time. In the long run, home bias remains large due to the interaction between information and portfolio choices. Empirical evidence supports the main predictions of our model, namely that bias increases with information capacity and decreases with …nancial openness.

The aim of this study was to demonstrate the occurrence of stochastic resonance (SR) in spinal and cortical potentials elicited by periodic tactile stimuli in the anaesthetised cat. The periodic tactile stimuli were applied on the central pad of the hindpaw and the noisy tactile stimuli on the glabrous skin of the third hindpaw digit. This protocol allowed that the signal and noise were mixed not in the skin but in the somatosensory regions of the central nervous system. The results show that a particular level of tactile noise can increase the amplitude of the spinal and cortical potentials elicited by periodic tactile stimuli. The topographical distribution of evoked potentials indicates that the effects of noise were spatially restricted. All cats showed distinct SR behavior at the spinal and cortical stages of the sensory encoding. Such SR was abolished in the cortical but not in the spinal recording after the sectioning of the ascending pathways. This suggests that the spinal neurones may also contribute to the SR observed at the cortical level. The present study documents the first evidence that the SR phenomenon occurs in the spinal and cortical somatosensory system itself and not only in the peripheral sensory receptors.

This thesis addresses two known quantities in quantum information science: (1) entanglement cost, and (2) Holevo capacity. These quantities will be crucial values when teleportation becomes common in daily life, perhaps centuries from now. Assume that Alice desires to send a singing Japanese cricket to her friend Bob in America, and that Alice and Bob already share a quantum entanglement. First, Alice sends Bob a mass of information bits resulting from the interaction between the cricket she holds in her hand and half of the entanglement. Subsequently, Bob receives the information bits and manipulates the other half of the entanglement, transforming them back into the original cricket. Examining this situation from an instrumental engineering viewpoint, quantifying the amount of the quantum entanglement and the number of information bits is crucial for this transmission. If both values are enough, Alice could even send herself to Bob's place instead of the tiny cricket. The topics of this thesis therefore are: (1) the mathematical properties of the entanglement cost, such as whether it is an additive measure similar to normal length or weight; and (2) how to calculate the Holevo capacity, an ultimately achievable limit of the information conveyance capacity of an information channel, such as of a single photon passing through an optical fiber or space. These two distinct quantities are magically tied together by several "additive or not" hypotheses, which await mathematical proof.

Capacity for information in biological molecules is traditionally thought to reside in the primary sequence of proteins and RNA, recorded in the DNA. With the exception of some RNA molecules, proteins, if not structural, carry their information in binding sites for substrates of reactions, or in binding sites for control molecules. Some proteins bind to complex carbohydrates in a carbohydrate-specific fashion, including enzymes, lectins and antibodies. These carbohydrates, assembled by sequential glycosyl transferases, also carry biological information, the other side of which is a binding protein that recognizes a specific sugar monosaccharides, sequence, anomerity, linkage, ring size, branching and substitution. It is the latter 7 parameters, however, that give carbohydrates a very large potential for information-carrying capacity in a short sequence. An exponentially growing body of knowledge exists in this aspect of carbohydrate function.

We present a simple model of quantum communication where a noisy quantum channel may benefit from the addition of further noise at the decoding stage. We demonstrate enhancement of the classical information capacity of an amplitude damping channel, with a predetermined detection threshold, by the addition of noise in the decoding measurement.

The maximum rates for information transmission through noisy quantum channels has primarily been developed for memoryless channels, where the noise on each transmitted state is treated as independent. Many real world communication channels experience noise which is modelled better by errors that are correlated between separate channel uses. In this paper, upper bounds on the classical information capacities of a class of quantum memory channels are derived. The class of channels consists of indecomposable quantum memory channels, a generalization of classical indecomposable finite-state channels.

The improved quantum scheduling algorithm proposed by Grover has been generalized using the generalized quantum search algorithm, in which a unitary operator replaces the Walsh-Hadamard transform, and π/2 phase rotations replace the selective inversions, in order to make the quantum scheduling algorithm suitable for more cases. Our scheme is realized on a nuclear magnetic resonance (NMR) quantum computer. Experimental results show a good agreement between theory and experiment.

Selective breeding is considered as a communication channel, in a novel way. The Shannon informational capacity of this channel is an upper limit on the amount of information that can be put into the genome by selection: this is a meaningful upper limit to the adaptive complexity of evolved organisms. We calculate the maximum adaptive complexity achievable for a given mutation rate for simple models of sexual and asexual reproduction. A new and surprising result is that, with sexual reproduction, the greatest adaptive complexity can be achieved with very long genomes, so long that genetic drift ensures that individual genetic elements are only weakly determined. Put another way, with sexual reproduction, the greatest adaptive complexity can in principle be obtained with genetic architectures that are, in a sense, error correcting codes. For asexual reproduction, for a given mutation rate, the achievable adaptive complexity is much less than for sexual reproduction, and depends only weakly on genome length. A possible implication of this result for genetic algorithms is that the greatest adaptive complexity is in principle achievable when genomes are so long that mutation prevents the population coming close to convergence.

An information capacity-cost analysis is used to find a minimum cell size for information capacity increase in cellular wireless network, thus a theoretical limit to cell size reduction. The results show that as the cell size radius R reduces to 300 m and less, the proposed interference model show a 5.76-18.89 % decrease in the information capacity per unit cost (£, $, etc) at microwave carrier frequencies f c > 3.35 GHz. This result illustrates that there is a theoretical limit to cell size reduction in relation to information capacity performance and cost. An inductive approach is used to generate a formula for calculating the number of co-channel interfering cells N n in a cellular wireless site layout. Such a formula allows one to calculate the number of co-channel interfering cells in subsequent tiers of a cellular wireless site layout. The geometric derivation shows that the number of co-channel interfering cell N n in a subsequent tier is the product of the number of co-channel interfering cells in the first tier N I and the tier number n. Thus, the number of co-channel interfering cell in a subsequent tier N n = N I × n. This formula enables subsequent tier co-channel interference to be included in the information capacity analysis of future and emerging, and finding the minimum cell size for information capacity increase in a cellular wireless communication system.

Two-Dimensional Magnetic Recording (TDMR) differs from traditional track recording in that the bit-size-tograin-size ratio is drastically reduced so that the channel bits are roughly of the same sizes as those of the magnetic media grains. This is envisioned to be achieved by shingled writing, a write-process in which the corner-writer partially overwrites the previously written adjacent track, effectively writing very closely spaced narrow tracks with no guard bands. Since the tracks are very narrow, they can be read by an array of read-elements whose spacing is equal to the narrow track pitch, creating a 2dimensional (2D) array of readback signals (the two dimensions being the cross-track and the down-track dimensions). In the absence of an array of read elements, the same 2D readback can be obtained by progressive scans of a single read element that reads one narrow track at a time and stores the readback signal in a 2D array. In this paper, we present simple magnetic grain media models and shingled-write process models that capture the essence of TDMR. We assume that the granular recording medium consists of randomly shaped tiles (each tile represents a grain), randomly covering the medium plane. We then derive a suitable 2-dimensional read/write process model. Using proper information-theoretic inequalities and bounding techniques, we derive methods to bound the capacities of TDMR channels. The results reveal that information capacities in excess of 0.6 user bits per grain are possible to attain over TDMR channels.

In this article, we explore the relationship between self-managed work teams and innovative behaviors they demonstrate from the socio-technical, social learning, and cognitive perspectives. Our specific argument is that self-managed work teams foster greater autonomy, increase communication among team members, and intensify their commitment to the team and the organization, which in turn enhances innovative behaviors in such teams. Propositions that may be of use to both the theoretician and the practitioner are presented.

Free Space Optics(FSO) Communication system quality can be improved by using Multiple Transmitters/Receivers. With the current needs of this technology for longer distance, the qualitative analysis of the system has become essential. In this work, the received power level (PR) and bit error rate (BER) are considered to determine the FSO link performance. And also the variation of the parameters considered along with the variation of the distance of the transmission. The relationship between the two parameters are investigated and analyzed. Furthermore, the received power for various number of TXs and RXs are experimentally measured and compared with the theoretical value calculated. The first part of the work deals with the variation in the number of transmitters and receivers. The second part is the variation of the distance. Finally, the total performance is analyzed.

A distribution of the longest distances to the nearest neighbour alongside a genetic sequence is investigated. It is shown the real DNA sequences differ basically from any model one, while the figures of the distribution prove almost lack of ergodicity in these latter.

Interactive hashing has featured as an essential ingredient in protocols realizing a large variety of cryptographic tasks, notably oblivious transfer in the bounded storage model. In interactive hashing, a sender transfers a bit string to a receiver such that two strings are received, the original string and a second string that appear to be chosen at random. This paper presents a self-contained, information theoretic study of interactive hashing. We start by formalizing the notion of interactive hashing as a cryptographic primitive, disentangling it from the specifics of its various implementations. To this end, we present an application-independent set of information theoretic conditions that all interactive hashing protocols must ideally satisfy. We then provide a detailed analysis of a standard implementation of interactive hashing which is shown to satisfy all the conditions of our definition. Our analysis represents a significant improvement over previous attempts in more restricted contexts. Despite its generality, it offers a considerably simpler proof of security. Moreover, it establishes a tighter upper bound on the cheating probability of a dishonest sender, who wishes to manipulate the protocol so that both output strings have some rare desirable property. In particular, we prove that if the set of desirable strings for the dishonest sender represents a fraction f of all strings, then the probability that both outputs will be from this set is no larger than 15.6805 • f. This upper bound is valid for any f and is tight up to a small constant, since a sender acting honestly would get two outputs from this set with probability very close to f. We illustrate the power of interactive hashing as a cryptographic tool by surveying protocols achieving oblivious transfer in the bounded storage model, which typically rely heavily on interactive hashing.

Several properties of the products of finite maximal prefix, maximal biprefix, semaphore, synchronous, maximal infix and maximal outfix codes are discussed respectively. We show that, for two nonempty subsets X and Y of A Ã such that the product XY being thin, if XY is a maximal biprefix code, then X and Y are maximal biprefix codes. Also, it is shown that, for two finite nonempty subsets X and Y of A Ã such that the product XY being unambiguous, if XY is a semaphore code then X and Y are semaphore codes. Finally, two open problems to the product of finite semaphore and maximal infix codes are presented.

Low information capacity interfaces (LICIs) such as binary switches, often become the only means by which people with disabilities can interact with their assistive devices. Although numerous studies have been done to develop novel ways to estimate user intention in cases of extreme disability, little attention has been paid to the process by which the acquired information is translated into specific control tasks (e.g. navigating a menu or positioning a cursor). It has been traditionally assumed that an exact estimate of user intention is needed to produce reliable control commands. When using LICIs, this requires lengthy synchronous control strategies which demand a significant amount of sensory feedback. In order to develop appropriate alternatives, we have redefined the latter as a communication, rather than a control problem. Thus, our objective becomes the efficient transmission of control commands from the user to the device through LICIs. By applying basic principles of information theory, we use the natural uncertainty of asynchronous human-machine interactions with LICIs to obtain increasingly accurate estimates of user intention. Our theory is based on the implementation of a virtual intention pointer (VIP) that maps the user's intention on the device domain. Thus, changes in the VIP's behavior result in corresponding changes to the device being controlled. The VIP's behavior is described by user intention rings (UIRs). UIRs are unidimensional, viscoelastic spaces, mapped on each of the angular dimensions of the VIP's velocity vector. A variety of kernels representing different communication/control strategies may be used to deform the UIRs creating nonparametric estimates of user intention. Although cursor positioning and trajectory following in 2-and 3-D spaces are ideal applications for this theory, it can also be used for virtual navigation in multidimensional domains to control, for instance, prosthetic or robotic arms with multiple (>3) degrees of freedom.

Global teams are an important work structure in software development projects. Managing such complex global software projects presents many challenges to traditional leadership wisdom, in particular, how, why and when the leaders should delegate responsibility and authority. Delegation is considered an important leadership component to motivate and grow subordinates. Cultural differences, skill level disparity and potential competition between different software development sites creates a management context which is much different from where traditional leadership theories were developed. This study investigates leader delegation behaviors in global software teams and explores the reasons and impact of delegation strategies on global team performance. Semistructured interviews and a survey was used to collect data from global software team managers and members from four countries of a Fortune 100 IT service company. The results of this study include in-depth analysis of hows and whys of leader delegation in global teams and a theoretical model for analyzing global team leader delegation occurrence and effects.

The aim of this study was to demonstrate the occurrence of stochastic resonance (SR) in spinal and cortical potentials elicited by periodic tactile stimuli in the anaesthetised cat. The periodic tactile stimuli were applied on the central pad of the hindpaw and the noisy tactile stimuli on the glabrous skin of the third hindpaw digit. This protocol allowed that the signal and noise were mixed not in the skin but in the somatosensory regions of the central nervous system. The results show that a particular level of tactile noise can increase the amplitude of the spinal and cortical potentials elicited by periodic tactile stimuli. The topographical distribution of evoked potentials indicates that the effects of noise were spatially restricted. All cats showed distinct SR behavior at the spinal and cortical stages of the sensory encoding. Such SR was abolished in the cortical but not in the spinal recording after the sectioning of the ascending pathways. This suggests that the spinal neurones may also contribute to the SR observed at the cortical level. The present study documents the first evidence that the SR phenomenon occurs in the spinal and cortical somatosensory system itself and not only in the peripheral sensory receptors.

We study the following combinatorial version of the Slepian-Wolf coding scheme. Two isolated Senders are given binary strings X and Y respectively; the length of each string is equal to n, and the Hamming distance between the strings is at most αn. The Senders compress their strings and communicate the results to the Receiver. Then the Receiver must reconstruct both strings X and Y. The aim is to minimize the lengths of the transmitted messages. For an asymmetric variant of this problem (where one of the Senders transmits the input string to the Receiver without compression) with deterministic encoding a nontrivial bound was found by A. Orlitsky and K. Viswanathany, [15]. In our paper we prove a new lower bound for the schemes with syndrome coding, where at least one of the Senders uses linear encoding of the input string. For the combinatorial Slepian-Wolf problem with randomized encoding the theoretical optimum of communication complexity was found in [24], though effective protocols with optimal lengths of messages remained unknown. We close this gap and present a polynomial time randomized protocol that achieves the optimal communication complexity.

Is visual attention mediated by a general-purpose processor with a small data capacity? Such an attentive processor could perform a wide range of transformations upon a small amount of image data. We suggest that this limited capacity corresponds to a fixed amount of information, measured in bits. We measure how much information an observer's attention can handle by measuring how much we can restrict display information without impairing the observer's performance. The attentive visual tasks we study are the detection of a stationary dot in a field of moving dots, and the detection of a static square in a field of flashing squares. Performance of these tasks is perfect up to a critical number of elements (the span of attention) and then falls as the number of elements increases beyond this critical number. The display information required for unimpaired performance in each of these tasks is low; the results indicate that visual attention pm only 30 to 60 bits of display information. Visual attention Span of attention Critical number Information capacity

In this paper, we deal with the design of high-rate multilevel two-dimensional (2D) bar codes for the print-andscan channel. Firstly, we derive an upper bound on the maximum achievable rate of these codes by studying an inter-symbol interference (ISI) free, perfectly synchronized, and noiseless print-and-scan channel, in which the printer device uses halftoning to simulate multiple gray levels. Secondly, we briefly review three state-of-the-art coded modulation techniques for the additive white Gaussian noise channel (AWGN) in the high signal-to-noise (SNR) ratio regime, namely, multilevel coding with multistage decoding (MLC/MSD), multilevel coding with parallel independent decoding (MLC/PID), and bit-interleaved coded modulation (BICM). Thirdly, we present a new model of the print-and-scan channel specifically adapted to the multilevel 2D bar code application. This model, inspired by our experimental work, assumes no ISI and perfect synchronization, but independence between the channel input and the noise is not supposed. We study the problem of finding the information capacity of our channel model and extend the theory of MLC/MSD to this type of channels. Finally, we present experimental results confirming the validity of our channel model, and showing that multilevel 2D bar codes using MLC/MSD can reliably achieve the high-rate storage requirements of many multimedia security and data management applications.

This deliverable (D6.2 Scalability of machine learning algorithms over every POMs) is the result of task T6.2 (Assessing scalability and computational efficiency of federated privacy-preserving machine learning algorithms) and describes the results obtained after intensive tests carried out to analyse the main technical characteristics of every algorithm over every POM. We have followed the Goal Question Metric methodology described in D6.1, and the main goals are to assess the scalability, computational efficiency and performance of the proposed and implemented schemes in the MUSKETEER Machine Learning Library (MMLL).

This paper explores the improvement in information capacity and practical data rate that is possible with adaptive antenna technology applied to wireless-multipath communication channels. Whereas the conventional view is that multipath-signal propagation is an impediment to reliable communication, this paper shows that multipath can actually multiply the achievable data rate for wireless channels provided that the appropriate communication structure is employed. Multivariate discrete multitone (MDMT) combined with multivariate trellis-coded modulation (MTCM) is proposed and analyzed as a practical means of realizing a multiplicative-rate advantage in the case where channelstate information is not available at the transmitter. In Rayleigh fading, these new techniques are shown to simultaneously achieve a multiplicative data-rate advantage and lower error rate as compared to conventional coded orthogonal-frequency division multiplexing. Optimal minimum mean square error (MMSE), adaptive MDMT channel-estimation techniques are derived. The effects of channel-estimation error on MTCM are analyzed.

Early physiological studies emphasized changes in the discharge rate of basal ganglia in the pathophysiology of Parkinson's disease (PD), whereas recent studies stressed the role of the abnormal oscillatory activity and neuronal synchronization of pallidal cells. However, human observations cast doubt on the synchronization hypothesis since increased synchronization may be an epi-phenomenon of the tremor or of independent oscillators with similar frequency. Here, we show that modern actor= critic models of the basal ganglia predict the emergence of synchronized activity in PD and that significant non-oscillatory and oscillatory correlations are found in MPTP primates. We conclude that the normal fluctuation of basal ganglia dopamine levels combined with local cortico-striatal learning rules lead to noncorrelated activity in the pallidum. Dopamine depletion, as in PD, results in correlated pallidal activity, and reduced information capacity. We therefore suggest that future deep brain stimulation (DBS) algorithms may be improved by desynchronizing pallidal activity.

We give an elementary self-contained proof that the minimal entropy output of arbitrary products of channels ρ → 1 d−1 ½−ρ T is additive. This paper is concerned with efficiency of transmission of classical information through a particular quantum channel. Generally, the minimal entropy output of a general quantum channel Γ, given in terms of a completely positive trace-preserving map on the d-dimensional density matrices, quantifies the minimal noise produced by the channel in terms of the von Neumann entropy of transformed pure states MEO Γ := inf ϕ, ϕ =1

This research asked 252 upper-, middle-, and first-line-level managers in organizations experiencing radical change to assess the effects of their own leaders' communications and behaviors on their perceptions of the change process. Results indicated that the frequency of exhibition of most behaviors by leaders positively affected subordinates' perceptions of change. For three types of behaviors, soliciting upward feedback, driving change, and providing resources, the importance of these behaviors to the subordinates' moderated perceptions of the change process. Discussion of these results and their implications conclude the study.

The unprecedented access offered by the World Wide Web brings with it the potential to gather huge amounts of data on human activities. Here we exploit this by using a toy model of financial markets, the Minority Game (MG), to investigate human speculative trading behaviour and information capacity. Hundreds of individuals have played a total of tens of thousands of game turns against computer-controlled agents in the Web-based Interactive Minority Game. The analytical understanding of the MG permits fine-tuning of the market situations encountered, allowing for investigation of human behaviour in a variety of controlled environments. In particular, our results indicate a transition in players' decision-making, as the markets become more difficult, between deductive behaviour making use of short-term trends in the market, and highly repetitive behaviour that ignores entirely the market history, yet outperforms random decision-making.

Mestre em Contabilidade pela FUCAPE. MBA em Finanças e Mercado de Capitais pelo IBMEC. Especialista em Comércio Exterior pela UFES. Graduado em Ciências Contábeis pela UFES. Atualmente é professor adjunto da Universidade Federal do Espírito Santo (UFES), lecionando para os cursos de graduação, pós-graduação e mestrado. Contato:

RESUMO Ball e Brown (1968) relatam que lucro líquido é uma informação contábil relevante para os investidores como critério na decisão de investir na empresa, deste modo, as informações produzidas pelas demonstrações contábeis podem explicar o retorno das ações. Assim, estudo tem por objetivo analisar a influência das políticas de dividendos e o nível de governança nos retornos anormais das ações das empresas listadas na BM&FBovespa no período de 2010 a 2013. Esta pesquisa é quantitativa, com análise de 255 empresas, sobre os impactos do anúncio do pagamento dos dividendos no retorno anormal da ação por meio de um modelo de regressão múltipla linear. Os resultados apontaram que os retornos anormais sofreram grandes variações e não há uma relação direta entre dividend yields e retorno anormal das ações, contrapondo o estudo de Novis Neto e Saito (2003). No entanto, corroboraram com Nagem e Amaral (2012), quanto aos retornos anormais sofrerem grandes variações sem uma relação direta entre dividend yields e retorno anormal das ações; a variável governança corporativa apresentou resultado insignificante, com evidências de que a mesma não influência no retorno anormal das ações; e empresas listadas em alguns níveis diferenciados de governança corporativa do Ibovespa não tem significância no poder de informatividade do retorno anormal.