Questions tagged [convolutional-neural-network]
Convolutional Neural Networks are a type of neural network in which only subsets of possible connections between layers exist to create overlapping regions. They are commonly used for visual tasks.
1,486 questions
1 vote
0 answers
37 views
How to solve the "regression dillution" in Neural Network prediction?
"Neural network regression dilution" refers to a problem where measurement error in the independent variables of a neural network regression model biases the sensitivity (of outputs to input ...
1 vote
0 answers
31 views
How does a VGG block encode relative position information?
Transformers either need the input vectors added with absolute position embedding (learned or sinusoidal positional encoding) or relative position encoding (like RoPE) or have the proximity bias term ...
1 vote
0 answers
47 views
Potential CNN Overfitting Due to Limited Training Data
Neural Network Beginner here. I am currently implementing a CNN on PyTorch for recognizing Japanese handwritten letters, which has 46 classes of outputs. I found a dataset on Kaggle https://www.kaggle....
0 votes
0 answers
65 views
Why is one-hot encoding used in RL instead of binary encoding?
Basically, the question above: in RL, people typically encode the state as a tensor consisting of a plane with "channels", i.e. original Alpha Zero paper. These channels are typically one-...
2 votes
0 answers
52 views
Should Feature Maps be considered for Dimensionality Reduction with Convolutional AutoEncoders
I am trying to use a convolutional autoencoder to perform dimensionality reduction with the ultimate goal of reconstructing temperature fields. As I understand it, the goal of an autoencoder is to ...
0 votes
0 answers
78 views
What is a suitable loss function for predicting cos(φ) and sin(φ) of a circular data using a CNN?
I want to predict an angular parameter ($\phi$) from some signal using a CNN. Due to the architecture of my code, the regression is done on the two targets ($\cos\phi$, $\sin\phi$). I created a model ...
0 votes
0 answers
96 views
Doubt regarding optimizers : Adam vs SGD
As a part of my undergraduate project , I was firstly given the task to benchmark a couple of standard CNN models (i)custom model with 4 conv layers (ii)ResNET18 on the FashionMNIST and CIFAR-10 ...
0 votes
0 answers
26 views
How to correlate inputs in multimodal neural nets (CNN & feedforward)
I am working on a Whole Slide Image (WSI) classifier, and for that task, CNNs (with transfer learning) have been producing average-to-decent results. However, for each WSI, I also have quantitative ...
0 votes
0 answers
48 views
How to treat NAs in CNN output which is an input to a NN
I have a dataframe of 5504 observations. For some of those 5504 observations, there are 965 images available. I want to add the information of the images to the information coming from the tabular ...
0 votes
0 answers
72 views
How to improve accuracy on a homemade plant classifier CNN?
I've got a CNN that I've written in Java. I've tested the code on MNIST and got 94% accuracy (not amazing, but evidence that the backpropagation works). However, I've now moved on to trying to get a ...
0 votes
0 answers
54 views
Understanding Backpropagation in Convolutional layer
I need help understanding backpropagation in the convolutional layer. From what I know so far, the forward phase is as follows: where, the tensor $A_{3\times3\times1}$ refers to the feature map in ...
4 votes
1 answer
97 views
Training Trajectory: Erratic behaviour of AUC and PR-AUC in CNN training
I have trained a CNN that aims to predict whether a patient died or not. The model inputs images of size 244x244 and outputs the binary outcome (which is highly imbalanced 7%). The dataset I have is ...
1 vote
0 answers
60 views
Overfitting problem in classification CNN
So I have a school project which is to train a CNN with our own architecture to be able to classify marine mammals with a minimum accuracy of 0.82 I have been trying a lot of things and different way ...
0 votes
0 answers
35 views
Can we set different patience in early stopping criteria based on improvement in validation accuracy of baseline and pruned model
I am doing unstructured feature weight pruning in a CNN. First I trained a baseline model without pruning and stopping criteria based on improvement in the validation accuracy and set the patience 15 ...
2 votes
1 answer
86 views
Question about number of filters in layer in CNN
I have a question about something that is written in the 100 page machine learning book, which is free here: https://themlbook.com/wiki/doku.php It is regarding chapter 6 regarding neural networks. ...