121,790 research outputs found
Generalization Learning in a Perceptron with Binary Synapses
No full textWe consider the generalization problem for a perceptron with binary synapses, implementing the Stochastic Belief-Propagation-Inspired (SBPI) learning algorithm which we proposed earlier, and perform a mean-field calculation to obtain a differential equation which describes the behaviour of the device in the limit of a large number of synapses N. We show that the solving time of SBPI is of order N√(logN) , while the similar, well-known clipped perceptron (CP) algorithm does not converge to a solution at all in the time frame we considered. The analysis gives some insight into the ongoing process and shows that, in this context, the SBPI algorithm is equivalent to a new, simpler algorithm, which only differs from the CP algorithm by the addition of a stochastic, unsupervised meta-plastic reinforcement process, whose rate of application must be less than 2/√(πN) for the learning to be achieved effectively. The analytical results are confirmed by simulations
Systematically and efficiently improving -means initialization by pairwise-nearest-neighbor smoothing
No full textWe present a meta-method for initializing (seeding) the -means clustering
algorithm called PNN-smoothing. It consists in splitting a given dataset into
random subsets, clustering each of them individually, and merging the
resulting clusterings with the pairwise-nearest-neighbor (PNN) method. It is a
meta-method in the sense that when clustering the individual subsets any
seeding algorithm can be used. If the computational complexity of that seeding
algorithm is linear in the size of the data and the number of clusters ,
PNN-smoothing is also almost linear with an appropriate choice of , and
quite competitive in practice. We show empirically, using several existing
seeding methods and testing on several synthetic and real datasets, that this
procedure results in systematically better costs. In particular, our method of
enhancing -means++ seeding proves superior in both effectiveness and speed
compared to the popular "greedy" -means++ variant. Our implementation is
publicly available at https://github.com/carlobaldassi/KMeansPNNSmoothing.jl.Comment: https://openreview.net/forum?id=FTtFAg3pek 16 pages (+8 appendix), 2
figures, 4 tables (+14 appendix). Transactions on Machine Learning Research,
Dec 202
Non-linear integration of crowded orientation signals
No full textAbstractCrowding of oriented signals has been explained as linear, compulsory averaging of the signals from target and flankers [Parkes, L., Lund, J., Angelucci, A., Solomon, J. A., & Morgan, M. (2001). Compulsory averaging of crowded orientation signals in human vision. Nature Neuroscience, 4(7), 739–744]. On the other hand, a comparable search task with sparse stimuli is well modeled by a ‘Signed–Max’ rule that integrates non-linearly local tilt estimates [Baldassi, S., & Verghese, P. (2002). Comparing integration rules in visual search. Journal of Vision, 2(8), 559–570], as reflected by the bimodality of the distributions of reported tilts in a magnitude matching task [Baldassi, S., Megna, N., & Burr, D. C. (2006). Visual clutter causes high-magnitude errors. PLoS Biology, 4(3), e56]. This study compares the two models in the context of crowding by using a magnitude matching task, to measure distributions of perceived target angles and a localization task, to probe the degree of access to local information. Response distributions were bimodal, implying uncertainty, only in the presence of abutting flankers. Localization of the target is relatively preserved but it quantitatively falls in between the predictions of the two models, possibly suggesting local averaging followed by a max operation. This challenges the notion of global averaging and suggests some conscious access to local orientation estimates
Input-driven unsupervised learning in recurrent neural networks
No full textUnderstanding the theoretical foundations of how memories are encoded and retrieved in neural populations is a central challenge in neuroscience. A popular theoretical scenario for modeling memory function is an attractor neural network with Hebbian learning (e.g. the Hopfield model). The model simplicity and the locality of the synaptic update rules come at the cost of a limited storage capacity, compared with the capacity achieved with supervised learning algorithms, whose biological plausibility is questionable. Here, we present an on-line learning rule for a recurrent neural network that achieves near-optimal performance without an explicit supervisory error signal and using only locally accessible information, and which is therefore biologically plausible. The fully connected network consists of excitatory units with plastic recurrent connections and non-plastic inhibitory feedback stabilizing the network dynamics; the patterns to be memorized are presented on-line as strong afferent currents, producing a bimodal distribution for the neuron synaptic inputs ('local fields'). Synapses corresponding to active inputs are modified as a function of the position of the local field with respect to three thresholds. Above the highest threshold, and below the lowest threshold, no plasticity occurs. In between these two thresholds, potentiation/depression occurs when the local field is above/below an intermediate threshold. An additional parameter of the model allows to trade storage capacity for robustness, i.e. increased size of the basins of attraction. We simulated a network of 1001 excitatory neurons implementing this rule and measured its storage capacity for different sizes of the basins of attraction: our results show that, for any given basin size, our network more than doubles the storage capacity, compared with a standard Hopfield network. Our learning rule is consistent with available experimental data documenting how plasticity depends on firing rate. It predicts that at high enough firing rates, no potentiation should occu
Reward sharpens orientation coding independently on attention
No full textRewarding improves performance. Is it due to modulations of the output modules of the neural systems or are there mechanisms favoring more 'generous' inputs? Some recent study included V1 in the the circuitry of reward-based modulations, but the effects of reward can easily be confused with effects of attention. Here we address this issue with a psychophysical dual task to control attention while orientation sensitivity on targets associated to different levels of reward is measured. We found that different reward rates improve orientation discrimination and sharpen the internal response distributions. Data are unaffected by changing attentional load nor by dissociating the feature of the reward cue from the feature relevant for the task. This suggests that reward may act independently on attention by modulating the activity of early sensory stages, perhaps V1, through a SNR improvement of task-relevant channels. Reward acts like attention, but using separate channels
A Multi-Language Comparison of Influences on Author Verification using Character N-Grams
No full textWe create a new multi-language corpus for author verification based on Wikipedia talkpages, and evaluate the influence that differences in topic and time have on character n-gram author profiles. Topic alignment between two texts is found to increase author verification precision, and an authors writing style is found to change over time, but not more significantly after 3 years than after 1 year.Information ArchitectureWISElectrical Engineering, Mathematics and Computer Scienc
Appropriate Similarity Measures for Author Cocitation Analysis
No full textWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
The vanishing author in computer-generated works: a critical analysis of recent Australian case law
No full textAbstract
The use of software is ubiquitous in the creation of many copyright works, yet the requirement in copyright law that every work have a human author who engages in independent intellectual effort means that its use may prevent copyright subsistence. Several recent Australian cases have refocused attention on authorship as an essential criterion of copyright subsistence, and these cases suggest that much computer-produced output may be authorless and thus lack copyright protection. This article, the first in a two-part series, analyses how each case deals with the question of authorship of computer-produced works and why the use of software diminishes copyright protection for a significant number of computer-generated works. The article critiques the application of conventional notions of human authorship developed in the pre-computer age to modern productions and suggests alternative approaches to authorship that satisfy both the major objectives of copyright policy and the need to adapt to the computer age. The article argues that, without a broader judicial approach to authorship of computer-generated works, Parliament must remedy the lacuna in protection for these ‘authorless’ works. Possible solutions for reform are suggested. In a forthcoming article, the author comprehensively examines those reform proposals
Diffusive author(s), cohesive author: Analysis of S/N (1994)
No full textThis study indicates the ways in which various aspects of the author(s) are brought forth in Dumb type’s performance art, the S/N production. Previous research has suggested a non-hierarchical organization of Dumb type and the absence of a “privileged author” in Dumb type’s collaborative work, S/N. However, the results that I have investigated from member’s interviews on the creative process of S/N along with my analysis of the recorded images of S/N, indicate a different aspect of the author(s). First, S/N was created through, so to speak, the collective ideas of the members of Dumb type. Further, S/N has at least nine quotations from previous performances, installations, and printed writings, besides the work-in-progress technique. Explicating one of the “author functions” as given by Michel Foucault, each text has plural subjects of the author. However, it has been revealed from members’ interviews that Teiji Furuhashi had a decision-making role in selecting the members’ ideas within the performance. Since then, S/N has had plural subjects of creation; however, Furuhashi is one of the subjects of creation along with the “privileged author.” S/N has plural authors (diffusive authors) yet at the same time, it has a “privileged author,” Teiji Furuhashi (cohesive author)
- …
