We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

Co-Planar Stereotaxic Atlas of the Human Brain—3-Dimensional Proportional System: An Approach to Cerebral Imaging, J. Talairach, P. Tournoux. Georg Thieme Verlag, New York (1988), 122 pp., 130 figs. DM 268

I have developed "tennis elbow" from lugging this book around the past four weeks, but it is worth the pain, the effort, and the aspirin. It is also worth the (relatively speaking) bargain price. Including appendixes, this book contains 894 pages of text. The entire panorama of the neural sciences is surveyed and examined, and it is comprehensive in its scope, from genomes to social behaviors. The editors explicitly state that the book is designed as "an introductory text for students of biology, behavior, and medicine," but it is hard to imagine any audience, interested in any fragment of neuroscience at any level of sophistication, that would not enjoy this book. The editors have done a masterful job of weaving together the biologic, the behavioral, and the clinical sciences into a single tapestry in which everyone from the molecular biologist to the practicing psychiatrist can find and appreciate his or

Principles of Neural Science

http://www.cnbc.pt/jpmatos/29.%20Bradley.pdf?origin=publicationDetail

Measuring emotion: The self-assessment manikin and the semantic differential

Well-being : the foundations of hedonic psychology

Measuring emotion: Behavior, feeling, and physiology

Hemodynamic and electrophysiological studies indicate differential brain response to emotionally arousing, compared to neutral, pictures. The time course and source distribution of electrocortical potentials in response to emotional stimuli, using a high-density electrode (129-sensor) array were examined here. Event-related potentials (ERPs) were recorded while participants viewed pleasant, neutral, and unpleasant pictures. ERP voltages were examined in six time intervals, roughly corresponding to P1, N1, early P3, late P3 and a slow wave window. Differential activity was found for emotional, compared to neutral, pictures at both of the P3 intervals, as well as enhancement of later posterior positivity. Source space projection was performed using a minimum norm procedure that estimates the source currents generating the extracranially measured electrical gradient. Sources of slow wave modulation were located in occipital and posterior parietal cortex, with a right-hemispheric dominance.

/pdf/large-scale-neural-correlates-of-affective-picture-2313uecv99.pdf

Large‐scale neural correlates of affective picture processing

The visual brain quickly sorted stimuli for emotional impact despite high-speed presentation ~ 3o r 5p er s !in a sustained, serial torrent of 700 complex pictures. Event-related potentials, recorded with a dense electrode array, showed selective discrimination of emotionally arousing stimuli from less affective content. Primary sources of this activation were over the occipital cortices, extending to right parietal cortex, suggesting a processing focus in the posterior visual system. Emotion discrimination was independent of formal pictorial properties ~color, brightness, spatial frequency, and complexity!. The data support the hypothesis of a very short-term conceptual memory store ~M. C. Potter, 1999!—shown here to include a fleeting but reliable assessment of affective meaning.

/pdf/fleeting-images-a-new-look-at-early-emotion-discrimination-3wrhjx1ud6.pdf

Fleeting images: a new look at early emotion discrimination.

Functional activation (measured with fMRI) in occipital cortex was more extensive when participants view pictures strongly related to primary motive states (i.e., victims of violent death, viewer-directed threat, and erotica). This functional activity was greater than that observed for less intense emotional (i.e., happy families or angry faces) or neutral images (i.e., household objects, neutral faces). Both the extent and strength of functional activity were related to the judged affective arousal of the different picture contents, and the same pattern of functional activation was present whether pictures were presented in color or in grayscale. It is suggested that more extensive visual system activation reflects "motivated attention," in which appetitive or defensive motivational engagement directs attention and facilitates perceptual processing of survival-relevant stimuli.

https://csea.phhp.ufl.edu/Publications/Old/Bradley2003.pdf

Activation of the visual cortex in motivated attention.

http://languagelog.ldc.upenn.edu/myl/ldc/llog.backup/Brizendine/Sabatinelli2005.pdf

Margaret M. Bradley

Papers

Measuring emotion: Behavior, feeling, and physiology

Large‐scale neural correlates of affective picture processing

Fleeting images: a new look at early emotion discrimination.

Activation of the visual cortex in motivated attention.

Parallel amygdala and inferotemporal activation reflect emotional intensity and fear relevance