Statistical significance of decoder prediction accuracy

a

Statistical significance of decoder prediction accuracy

across all scenes was determined using a Wilcox rank-sum test comparing the distribution of decoder prediction accuracies to a null distribution of prediction accuracies. For Epigenetic inhibitor more details, see Supplemental Experimental Procedures 13. Using the category probabilities predicted by the decoder for each scene in the validation set, we repeatedly picked from the 850 objects comprising the object vocabulary for the 20 best scene categories identified across subjects. Each object was picked by first drawing a category index with probability defined by the decoded scene category probabilities, followed by picking an object label with probability defined by the learned LDA model parameters. The learned LDA model parameters capture the statistical correlations of the objects in the learning database. Thus, the frequency of an object being picked also obeyed this correlation. The frequency distribution resulting from 10,000 BMS-907351 order independent object

label picks was then normalized. The result defined an estimated distribution of occurrence probabilities for the objects in the vocabulary. Statistical significance of object decoding accuracy across all scenes was determined using a Wilcox rank-sum test comparing the distribution of likelihood ratios for the decoder to a null distribution of likelihood ratios. For more details on this issue, see Supplemental Experimental Procedures 14. This work was supported by grants to J.L.G. from the National Eye Institute (EY019684), the National Institute of Mental Health (MH66990, and the National Science Foundation Center for the Science of Information (CCF-0939370).

We thank An Vu for data collection assistance and Tom Griffiths, Shinji Nishimoto, Tolga Cukur, Mark Lescoarte, Michael Oliver, Alex Huth, James Gao, Natalia Bilenko, Anwar Nunez, Ben Dichter, and Melanie Miller for helpful Rutecarpine discussions and comments. “
“One of the key mechanisms underlying the development of neuronal synapses and connectivity in the nervous system is synaptic adhesion. Synaptic adhesion molecules are thought to regulate diverse steps of synaptic development, including the formation, maturation, maintenance, and plasticity of neuronal synapses. A well-known example of such synaptic adhesion proteins are the neuroligins, which trans-synaptically interact with presynaptic neurexins ( Südhof, 2008). A series of studies on this interaction has significantly contributed to our current understanding of the mechanisms underlying synapse development, synaptic transmission, and plasticity and of neuropsychiatric diseases such as autism spectrum disorders. However, given the enormous diversity of neuronal connectivity in the nervous system, it is not surprising that a large number of synaptic adhesion molecules exist.

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>