Manifold/subspace for neural network parameters?
I have come across a lot of work that tries to exploit the fact that data (images/speech) lie on a common manifold. This has led to a lot of work in the area of trying to learn classifier parameters...
View ArticleZero-Shot Learning Through Cross-Modal Transfer (2013) [PDF]
submitted by galapag0 [link][1 comment]
View ArticleHelp with Theano MLP Tutorial
I keep receiving the error x: test_set_x[index * batch_size:(index + 1) * batch_size], IndexError: invalid sliceMy thinking is that the values being used to slice the numpy array are theano variables...
View ArticleClik here to view.
Clik here to view.
Cross social network analysis for building quality data sets
submitted by hoykg [link][comment]
View ArticleState-of-the-art for object tracking in videos
I looked around, but didn't really got a clear view of what works best for object tracking in videos.I know some methods with binary masks in deep learning for object detection in images. But videos...
View ArticlePower Law Effect with Pitman-Yor Process
I know about PYP and that they produce power law effect and am looking at some of the graphs, like:http://deliveryimages.acm.org/10.1145/1900000/1897842/figs/f1.jpgand wondering how would you draw the...
View ArticleAdvice for my ML project?
I'm in my final year of CS university and want to have a final year project related to ML. I thought about creating an app that scan a table from a paper or other physical support, and create an excel...
View ArticleWhat are some of the state of the art techniques for emotion recognition from...
submitted by shahj [link][comment]
View Article"Humanitarian" work within ML?
I am a graduate student in computer science with a mathematical background. I am considering what to write about for my master's thesis, knowing that my choice will determine to some extent what I'll...
View ArticleLooking for ML research opportunities at universities
Hey everyone,I am planning to pursue a research internship and a MS in Machine Learning.I have 3yrs experience working as a developer (2.5 yrs of that as a remote dev for a NY firm). I have developed...
View ArticleCould I use the If-Then-Else expression for surveys in social science?
Hi guys, I'm a researcher from Berlin and I'm interested in seeing how questionnaires and surveys that everyone uses in social science could improved/standardized by using the if-then-else format. The...
View ArticleA Comparison of Dropout and Weight Decay for Regularizing Deep Neural Networks
submitted by galapag0 [link][6 comments]
View ArticleNeural net input layer representations
How should I represent my neural net input, if each input node has 3 possible values?For example, a position on a Go board may be empty, have a white stone, or have a black stone.Should I use empty =...
View ArticleClassifying sounds
Hi everyone,What would be the best way to classify sounds? I only need to know if a sound is of one particular kind or not, e.g. drums or not, so slamming a locker and gunshots should not be in the...
View ArticleWhat are the implications that would apply to Machine Learning if P=NP?
The search for P=NP (or P!=NP) is not really making any headway, but I'm wondering how machine learning would be affected if someday we discover that P=NP. Would some learning be made faster? Would...
View ArticleWhat are current state of the art data clustering algorithms?
As we all know, clustering consists of a plethora of algorithms, eg. prototype algorithms, density-based, graph-theoretic, hierarchical, mixture models, and so on. I believe it's the data that usually...
View ArticleWhat does the No Free Lunch Theorem tell us?
People often reference the NFLT when talking about machine learning algorithms, especially when talking about whether one technique is better than another.I can't quite understand the implications, it...
View ArticleIs it possible to perform sparse filtering for RNNs?
Is it possible to perform sparse filtering ( http://ai.stanford.edu/~ang/papers/nips11-SparseFiltering.pdf )for recurrent neural networks? What other unsupervised methods for feature creation using...
View ArticleHow to design a multilabel classification layer in neural networks?
In ordinary classification a softmax may be used, where the maximum scoring class may be chosen. In case of multilabel classification how do we know how many labels ought to be chosen, when all the...
View Article