Questions tagged [neural-network]

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 designer having had a model of a real system.

Filter by
Sorted by
Tagged with
1
vote
0answers
19 views

What is correct equation for LR decision boundary?

I read that the equation perceptron decision boundary is given as follows:$$w^Tx-w_0=0$$ This can be proven as follows: Assuming $w$ is a unit vector (as we can multiply above equation with a ...
0
votes
0answers
7 views

Neural network with variable number of inputs

I have the following problem: I have a set of time-stamped articles, and labels for particular instants of time. I want to train a neural network such that it can learn which articles to take in as ...
0
votes
2answers
37 views

Logic behind pre-trained weights and transfer learning

I am not sure about the logic behind, how pre-trained weights actually make sense and translate into a new problem. To be more specific; for example in a object detection network, how would a model's ...
8
votes
3answers
2k views

Loss Function for Probability Regression

I am trying to predict a probability with a neural network, but having trouble figuring out which loss function is best. Cross entropy was my first thought, but other resources always talk about it in ...
2
votes
1answer
39 views

Is data subsampling appropriate for hyperparameter optimisation?

Fundamentally, under what circumstance is it reasonable to do HPO only on a subsample of the training set? I am using Population Based Training to optimise hparameters for a sequence model. My dataset ...
0
votes
0answers
6 views

Why concatenating these layers, why applying masks over and over to partial convoluted image?

I have to ask some questions about one topic. In this sentence of Nvidia's article of : https://arxiv.org/abs/1804.07723 , they are saying:"The last partial convolution layer’s input will ...
1
vote
0answers
14 views

How to feed the model with a stack of images instead of one by one?

I built a 2D model, but the dataset contains a group of images from different viewpoints for each patient, so the input should be a stack of images for each patient. I have compressed each group of ...
0
votes
2answers
90 views

Deep Learning Network decreasing in accuracy

In order to familiarize myself with semantic segmentation and convolutional neural networks I am going through this tutorial by MathWorks: Semantic Segmentation Using Deep Learning I did not use the ...
1
vote
2answers
149 views

Metrics values are equal while training and testing a model

I'm working on a neural network model with python using Keras with TensorFlow backend. Dataset contains two sequences with a result which can be 1 or 0 and positives to negatives ratio in dataset is 1 ...
7
votes
2answers
148 views

Neural Network Architecture for Identifying Image Copies

I have a large image collection and wish to identify the images within that collection that appear to copy other images from the collection. To give you a sense of the kinds of image pairs that I ...
0
votes
1answer
2k views

Multiple-input multiple-output CNN with custom loss function

I have a set of 2D input arrays $(n\times m)$ namely $A,B,C$ and I would like to predict two 2D output arrays namely $d,e$ for which I have the expected values. You can think of the inputs/outputs as ...
1
vote
1answer
148 views

What program to use to visualise neural network diagram and math functions

I am writing a paper about machine learning and I need to create some neural network diagrams and basic math functions I am describing. I need a program to create visually decent technical picture ...
1
vote
1answer
151 views

High Correlation between inputs of neural network

I am building a neural network to predict if a video is pornographic or not by analysing the bytes of upload and download at every 0.1 seconds for a total of 25.6 seconds. So, I have 512 input ...
2
votes
2answers
52 views

Modeling social media post scheduling optimization

Problem: I want to maximize performance for social media posts by optimizing the time when they are published. Current model: ...
1
vote
0answers
11 views

Time Series Forecasting with LSTMs in keras - convergence problem

I am trying to forecast a time series with multivariate input and multi output (multi step forecast). Since some of my input features are known for future time steps, wheras others are not, naturally ...
0
votes
1answer
20 views

How to improve regression neural network?

I am new to deep learning and data science and trying to increase my knowledge by working on some hackathons. Currently, the hackathon project I am working on has the task to predict the closing price ...
0
votes
0answers
6 views

Continuous Bag of Words loss function and training objective

CBOW from what I understand, obtains a probability distribution $P(w|c)$ for all words $w$ in the vocabulary, given context $c$. Th loss function is: $-logP(w|c)$, which means this would be maximised ...
4
votes
2answers
81 views

What are the reasons for drawing initial neural network weights from the Gaussian distribution?

Are there theoretical or empirical reasons for drawing initial weights of a multilayer perceptron from a Gaussian rather than from, say, a Cauchy distribution?
1
vote
2answers
1k views

Feature scaling for MLP neural network sklearn

I am working with a dataset where the features have multiple scales. Before running scikit-learns's MLP neural network I was reading around and found a variety of different opinions for feature ...
0
votes
0answers
14 views

Neural Network Design Intuition [closed]

What are the different motivations or designing principles for various neural networks? For example one layer neural network I can understand that tries to find the correlation between the target and ...
1
vote
0answers
11 views

Are less training epochs better in the following scenario

So I have a scenario in which the training data is being generated in response to what the Neural Network backed actor is doing. In essence its giving feedback to the Neural Network based on all of ...
0
votes
0answers
8 views

No gradients provided for any variable, when using Lambda to round model output

I have a problem where I need to predict some integers from an image. The problem is that this includes some negative integers too. I have done some reasearch and came accross Poisson which does count ...
1
vote
2answers
156 views

Designing a pretrained DNN for image similarity

I am pretty new to deep learning and really hope that you can help me. I want to write a python program that lets me choose an area in a reference image. This subimage of variable size should then be ...
0
votes
0answers
14 views

How to improve my deep LSTM model for time series?

I want to train a deep model for my time series power consumption dataset. I have created a model consist of CNN, BILSTM, Encoder-Decoder, and dense layers. here is my model: ...
3
votes
3answers
15k views

ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type int) in Python

I have written the following code for a neural network to perform regression on a dataset, but I am getting a ValueError. I have looked up to different answers and ...
2
votes
1answer
143 views

Do batch GD and stochastic GD give the same results?

If a neural network is trained on a dataset of M samples for N epochs, do batch GD and SGD give the same result? Is SGD is faster because utilize the hardware better? I am asking because I figured out ...
1
vote
0answers
15 views

Where Does the Normal Glorot Initialization Come from?

The famous Glorot initialization is described first in the paper Understanding the difficulty of training deep feedforward neural networks. In this paper, they derive the following uniform ...
0
votes
0answers
12 views

What does it mean when accuracy of regularized model is higher for training set than for validation set?

Accuracy of my regularized model is higher for training set than for validation set. The situation improves when regularization coeefficient is reduced: What does this really imply? From my ...
3
votes
1answer
225 views

Data normalization in nonstationary data classification with Learn++.NSE based on MLP

I need to predict technical aggregate condition using vibration monitoring data. We consider this data to be nonstationary i.e. distribution parameters and descriptive statistics are not constant. I ...
7
votes
2answers
200 views

How to implement hierarchical labeling classification?

I am currently working on the task of eCommerce product name classification, so I have categories and subcategories in product data. I noticed that using subcategories as labels delivers worse results ...
2
votes
1answer
47 views

Recursive Transfer Learning

Is there any methodology called Recursive Transfer Learning? For example, let's consider a situation that we have a lack of data while training a convolution neural network (CNN) for object detection ...
0
votes
1answer
55 views

Does adding of many FC layers during re-training increase the model size ? Are there any ways to optimize the size of model?

I am re-training a pretrained model VGG16. In the last layers, im using two FC layers of size 2048 each, with dropout=0.5. When I saved the model, the size of the ...
1
vote
1answer
304 views

Problem when cherry picking actions - Proximal Policy Optimization

I am using the implementation of PPO2 in stable-baselines (a fork of OpenAI's baselines) for a Reinforcement Learning problem. My observation space is $9x9x191$ and my action space is $144$. Given a ...
1
vote
1answer
40 views

Autoencoder not learning walk forward image transformation

I have a series of 15 frames with (60 rows x 50 columns). Over the course of those 15 frames, the moon moves from the top left to the bottom right. Data = https://github.com/aiqc/AIQC/tree/main/...
2
votes
2answers
81 views

Is a neural network able to learn to map a completely different feature vector to the same class

Is a neural network (for example a MLPClassifier in Python) able to learn to map a completely (or very) different input feature set to the same output class? Or is it better to work in this case with ...
1
vote
1answer
79 views

Recurrent Neural Networks Over Multiple Documents Over Time

So in my head, I have an idea about what this architecture should look like, or at least behave, but I am having trouble implementing it. So let me describe the problem, and if anyone has an idea on ...
0
votes
0answers
15 views

Data augmentation within epochs vs across epochs

Usually in deep learning data augmentation is applied by creating a new augmented version of each training sample for each epoch. Therefore the amount of training samples for each epoch stays the same ...
0
votes
1answer
36 views

Question About Discriminator of CycleGan

The Discriminator of CycleGan outputs not just a single value to say that the image is real or fake.... But It outputs a grid of numbers (like 8X8 or 7x7), where each number says whether one patch of ...
0
votes
0answers
12 views

How to create custom stochastic layer in tensorflow 2.0?

Recently, I have been looking into Stochastic neural networks and would like to try creating one; however, I am not sure where to start. I have experience in Python and have been learning TensorFlow ...
2
votes
1answer
239 views

Target Variable Encoding for Time Series Change point detection

I am working on a time series data for which I intend to impliment machine learning model for detecting change point in time series data. This data is recorded fom machinary and we have to predict ...
0
votes
0answers
10 views

Can we have neural network emulate XOR logic gate with single neuron in the hidden layer?

I came across following neural networks emulating logical XOR gate: Approach 1: Approach 2: But today, I came across below one: I dont get how this behaves as XOR, especially what does those ...
2
votes
1answer
392 views

Architecture for multivariate multi-time-series model where some features are TS specific and some features are global

I'm looking to build a time series model (using a TCN or a LSTM) with $N$ different series, each of which has $P$ series-specific features $\mathbf{X}$. My input array is of dimension $N \times t \...
1
vote
1answer
80 views

Cable angle measurement (rotation)

I need to detect the rotation of a cable (degree) in the x-axis with high precision [0.2 (or more) degree detection] from its original state. Detailed description: I have a cable that is set in its ...
6
votes
1answer
5k views

Is it possible to customize the activation function in scikit-learn's MLPClassifier?

Scikit-learn lists these as the implemented activation functions for it's multi-layer perceptron classifier: ...
2
votes
1answer
304 views

Similarity coloring of a self organizing map

I have implemented the algorithm to train self organizing maps in Python and it seems to be working well. I checked with some labeled data and the maps are learning the topology well. Here are some ...
1
vote
0answers
950 views

How to change learning rate of MomentumOptimizer in tensorflow

I am trying to implement VGG-16 architecture in TensorFlow. As mentioned in the paper, they changed the learning rate 3 time during their 74 epochs of training. ...
2
votes
1answer
23 views

Learning a board game using a genetic neural network

I've never really done any practical machine learning, this is just a hobby for me. I'm trying to create a process using a neural network to learn the board game "7 Wonders." Here's how I ...
1
vote
1answer
528 views

Accuracy and Loss in MLP

I am trying to explore models for predicting whether a team will win or lose based on features about the team and their opponent. My training data is 15k samples with 760 numerical features. Each ...
3
votes
2answers
268 views

Reward function to avoid illegal actions, minimize legal action and learn to win - Reinforcement Learning

I'm currently implementing PPO for a game with the following characteristics: Observation space: 9x9x(>150) Action space: 144 In a given state, only a handful of actions (~1-10) are legal The state ...

1
2 3 4 5
78