Artificial neural networks (ANN), are composed of 'neurons' - programming constructs that mimic the properties of biological neurons. A set of weighted connections between the neurons allows information to propagate through the network to solve artificial intelligence problems without the network ...

learn more… | top users | synonyms

1
vote
0answers
19 views

If you know the central moments of the data $X$, find a function $f$ for which $f(X)$ has arbitrary central moments

Say you are given one-dimensional data $X$, with mean $\mu$ and central moments $a_n$ which you know. Can you construct a function $f(x)$ which transforms the data such that $f(X)$ has the central ...
0
votes
0answers
10 views

Reshaping Inputs that contain continuous and discrete values

The inputs I am using are 2xN, where the first 1xN row is a continuous number, and the second 1xN row takes on only discrete values (that encodes a . I expect there to be a relation between A[1,i] and ...
0
votes
0answers
9 views

Unsupervised deep learning image clustering

I'm interested in trying to apply my (very) limited understanding of neural networks to a problem of image clustering. I realise that NN's are mainly used for regression or classification problems ...
0
votes
2answers
23 views

Division by zero with cross entropy cost function

I am using a tanh as my activation function for my NN. I also was using the cross entropy cost function previously when I had sigmoid neurons. The sigmoid neurons can never make it to zero but a ...
0
votes
0answers
14 views

Predicting value with MLP neural network

I am trying to predict a value using feedforward and back-propagation network but I get incorrect predicted value. I checked the gradients calculation and error is 1.79684799923e-11 < 1e-9 so ...
-1
votes
0answers
20 views

Can any arbitrary function be approximated by a rational function? [on hold]

To add a little more clarity to the title, can all functions be approximated by some rational function within a particular domain? I'm interested in the potential applications for machine learning. ...
1
vote
1answer
33 views

Time Series forecasting with useful predictor variables

I am playing with time series data related to a issue ticketing system. The system logs all open tickets at any one point and my task is to predict what the volume of open tickets will be in 5,10,15 ...
0
votes
1answer
25 views

Predict moon shape clusters with Keras NN: why stays my model non-linear?

I first did this tutorial to understand NN: http://www.wildml.com/2015/09/implementing-a-neural-network-from-scratch/ Now I try to rebuild it with Keras. This is the code: ...
3
votes
1answer
45 views

Transition from “old-school” neural network methods to deep learning?

As far I know the current state of deep learning favours a rather simplistic setup -- in short: many layers to allow for representational learning, maxout or a similarly suited activation function to ...
0
votes
1answer
33 views

Neural network can only follow increasing function

I am trying to program a simple neural network using python. For some reason my code is only working on functions which are increasing. The network I am using has 1 input, 1 output, and 2 layers of ...
0
votes
0answers
14 views

Variable-length time series - neural network

I have data about patient purchases - specifically, how late their fulfillment of prescriptions are, or if they filled them at all. I want to feed this data into a neural network to classify them. ...
0
votes
0answers
18 views

How to think about the architecture of the Convolutional Neural Network?

Recently, I've started to learn more about CNNs to use them in some computer vision tasks. At the moment, I have roughly good knowledge about different parts of a CNN such as layers, solvers, loss ...
0
votes
0answers
28 views

Stuck while trying to predict on data based on H2O Deep Learning model

I have created an H2O Deep Learning model in R for multi-class classification and I want to use it to perform prediction. I would have assumed that if I use the model to predict on the validation ...
0
votes
0answers
18 views

Support to interpret Performance Metrics from H2O Deep Learning model

Using H2O in R, when I perform the call h2o.performance(m1, dataframe.valid.H2O), the below metrics table is returned. Where can I find explanation of what each row ...
0
votes
0answers
22 views

Support to interpret classification charts from H2O DEEP LEARNING

I have executed H2O Deep Learning (with grid search) on my data-set for multi-class classification and the resulting charts from the first and second best model do not make much sense to me. It seems ...
0
votes
0answers
6 views

How to use LSUV initialization in tensorflow?

How to use LSUV initialization in tensorflow? Tensorflow variable seems only have those few initializer.
0
votes
0answers
22 views

Neural networks that can reach state-of-the-art accuracy with two or three hours training?

Are there some neural networks that can reach state-of-the-art accuracy with two or three hours training, on dataset like CIFAR, MNIST,etc...
0
votes
0answers
8 views

Train softmax temperature during function optimisation

Instead of simply choosing a softmax temperature, does anyone include this parameter during function optimisation? e.g. in the last layer of a neural network during backprop? Does it make sense to ...
4
votes
0answers
39 views

Recurrent Neural Network (RNN) topology: why always fully-connected?

I've started reading about Recurrent Neural Networks (RNNs) and Long Short Term Memory (LSTM) ...(...oh, not enough rep points here to list references...) One thing I don't get: It always seems that ...
3
votes
1answer
56 views

Is it reasonable to study neural networks without mathematical education?

Given the modern state of machine learning technologies and tools (e.g. TensorFlow, Theano, etc.), it seems like entry threshold have recently lowered and it is enough to be able to program on, say, ...
1
vote
1answer
28 views

Mixture Density Network: What is C?

I'm currently trying to implement a Mixture Density Network (MDN) based off of the original paper here. Most of the equations seem pretty straight forward but on page 6 (7 of the PDF) equation 23 has ...
0
votes
0answers
10 views

Neural Networks in Image Processing - Literature Reviews

I'm looking for good Literature Reviews on the use of Neural Networks in "Image Processing/Image Retrieval/Image Classification" and generally anything Image Related... Has some work been done in ...
0
votes
1answer
24 views

nnet package - is it neccessary to scale data?

Could you tell me is it necessary to scale data ? I am using this package for prediction. Not scaled data give the same efficiency of prediction as scaled data.
1
vote
1answer
10 views

Delimitation: feed forward- and radial basis networks

intro: I am trying to get myself involved with the topic of neural networks for the purpose of a GPGPU project at university. Now, when reading on wikipedia two network types are the "Feed forward ...
0
votes
0answers
8 views

Dell statistica 13 prediciton with neural network [closed]

Can somebody explain to me does software Statistica ver. 13 use test data to stop traning neural networks (so the network is not underfitting or overfitting) or that are validations data for? I ask ...
0
votes
0answers
7 views

How to assign defined training set, val set and test set for training a Neural net in NNtoolbox?

To find an optimal number of hidden neurons and layers in my code using feedforward net, I use cross validation technique and cvpartition function to split data. Now my aim is to use this split data ...
0
votes
1answer
7 views

Training a RNN on time series: How to cope with different sequence origins?

I am pondering if I should apply a recurrent neural network on my data. Data is EEG from sleep, and thus there is much information hidden in the temporal domain. Ergo, RNNs make sense. Intro: I have ...
0
votes
1answer
27 views

How and what do I train in my Convolutional Neural Network [closed]

I have been trying to research and implement a convolution neural network in c++, and I think I understand the basic architecture of it. My problem is that I am incredibly confused as to what is ...
1
vote
0answers
21 views

Flexible prediction with neural network or other method

I want to use neural network for my first time, but I need to check if it fits for my case. So, my idea is to teach a model on data like Y = f(X1, X2...XN) and ...
0
votes
0answers
24 views

Testing dropout implementation

I have implemented dropout but on my particular task it does not seem to give any improvement so I am having a little doubts about the correctness of my implementation. Is there a reliable way to ...
1
vote
1answer
30 views

deep learning in mobile apps [closed]

Can deep learning be applied in mobile apps, and if possible how? Is it possible ? to deep learning needs more computational cost? how can we minimize the computational cost for deep learning on ...
0
votes
0answers
11 views

adjusting nnet model for prediction

Could someone give some hints how to adjust paramters in nnet model for predictions ? I mean following parameters: maxit, range, decay, size, and ...
0
votes
0answers
11 views

How to train LSTM for a simplest function recognition

I'm learning LSTM networks and decided to try synthetic test. I want LSTM network fed by some points (x,y) to distinguish between three basic functions: line: y = k*x + b parabola: y = k*x^2 + b ...
0
votes
1answer
30 views

Can we use network neural for nonumeric data?

I am going to use network net package to do predidction. Tell me please: Is it neccessary to do scaling of data ? Is it neccessary to convert all data into numeric/int type ? I am newbie at this ...
0
votes
0answers
9 views

Denoising Autoencoders pretraining for reLU network?

I am training a supervised deep neural network (at least 3 hidden layers) using reLU as activaton functions in the hidden layers. I read that for deep architectures pre-training can be beneficial and ...
0
votes
0answers
4 views

Activation of different states in a neural network

hello i want to ask if it is possible to create a neural network in mat lab that is able to have at least 4 states (by states mean something happening parallely not the output )which gets activated ...
0
votes
1answer
26 views

Neural Net Matrix Multiplication

I'm trying to figure out the matrix multiplications for the implementation of a single hidden layer neural net for MNIST digit recognition in Python. Like the following: ...
0
votes
1answer
21 views

Difference between MLP(Multi-layer Perceptron) and Neural Networks?

I am wondering about the differences. Based on my understanding, MLP is one kind of neural networks, where the activation function is sigmoid, and error term is cross-entropy(logistics) error. Looking ...
1
vote
0answers
6 views

Garson's algorithm for fully connected LSTMs

Garson proposed an algorithm, later modified by Goh (1995) for determining the relative importance of an input node to a network. In the case of a single layer of hidden units, the equation is $$ ...
2
votes
1answer
23 views

Neuron saturation occurs only in last layer or all layers?

In Chapter 3 of the Neural Networks and Deep Learning book, the text repeatedly states that neuron saturation depends only on the activation function of the output layer and the cost function, such ...
0
votes
1answer
15 views

Training CNN: Why, sometimes, does my CNN training go horribly wrong?

I have a CNN implemented on TensorFlow. I am using it on the MNIST dataset. 9/10 times, when I start training it, it very quickly gets up past 90%. However, once in a while, it zeros in on 9% ...
0
votes
1answer
20 views

Training of Hopfield network in Matlab

I have a matrix 35x5 with -1/1 representing letters (one column one letter). I present the network the matrix as attractors. After, I flip 3 pixels to see if the net is able to recall the correct ...
0
votes
0answers
16 views

Count branch points of a tree

Suppose I have a large b/w image of a tree and would like to find the branch points. To that end I generate, say, 1000 artificial 32x32 images of branches, where by construction I know which images ...
0
votes
0answers
12 views

Useful inputs from files for neural network?

I'm building a visual neural network sandbox. ATM i'm working on the Input nodes, where you have a browser to choose a file from your drives and drop it in the workspace. The workspace detects what ...
2
votes
1answer
53 views

Understanding Neural Networks Dropout - Visible vs. Hidden

I am trying to understand exactly how the dropout method typically works. I have been looking at the original Hinton paper but I can't seem to pull out this final detail. If I understand correctly, ...
1
vote
0answers
11 views

What can I infer about my problem domain/input data from the hyperparameters producing the optimum network configuration?

I am new to neural networks and am trying to solve a binary classification problem with a neural network. I tried network configurations with 1 to 6 hidden layers, and 1-50 neurons per layer. The ...
0
votes
0answers
56 views

Problem Training an LSTM network in Lasagne for simple task (determining parity of bit sequence)

I have been trying to gain some familiarity with the Lasagne libraries for machine learning, specifically LSTMs so I set up the following toy problem to determine the parity of a sequence of bits ...
0
votes
1answer
17 views

How well should backpropagation agree with finite difference methods when calculating derivatives of the error function?

I have attempted to write a Neural Network code, and it was suggested in my textbook (Bishop - Pattern Recognition & Machine Learning) that a very useful debugging technique is to check your ...
1
vote
0answers
28 views

Pros and Cons: LDA vs Neural Networks

LDA is an older approach for word representations, there are newer methods now like CBOW and Skip-gram. But what are the improvements of these models? Do they improve in every way or does LDA still ...
0
votes
1answer
24 views

How to Choose Activation Functions in a Regression Neural Network?

I'm having difficulties with some basics regarding the application of feed forward neural networks for regression. To be specific, lets say that I have an input variable $x \in \mathbb R^4$ and data ...