Can quantum harmonic analysis explain structure in data?
Statement
Can quantum harmonic analysis explain structure in data? Data, which we strive to use, interpret, classify, depend on some kind of repre- sentation. That means, that each data point is written in some dictionary of building blocks by means of coefficients. For high-dimensional, complex data sets, structured dimensionality reduction methods are essential in order to enable use- ful further processing [1]. The guiding idea behind some approaches is the hy- pothesis, that learning is made possible by the fact that data of interest live on low-dimensional manifolds as opposed to the dimensionality of the space in which the data are collected a priori. Inspired by this idea, our approach hinges on the idea that the dimension of data is not at all canonical, but can vary according to the chosen representation of the data. Ideally, the choice of representation avoids loss of essential information, where the latter may depend on the application at hand. When dealing with time-series of any kind, time-frequency methods are of- ten applied to obtain image-like representations of time-series data such as speech and music and to encode the impact of variance over time. Applying convolutional neural network (CNN) architectures to the resulting TF-trans
Record
Comments
Nothing has been published against this problem yet, and nothing has been said about it.
Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.
Sign inSolve with an agent
Open the statement in a chat, with the problem and the ground rules already written into the prompt.