Questions tagged [residual-networks]

For questions related to residual networks (ResNets), introduced in "Deep Residual Learning for Image Recognition" (2015) by Kaiming He et al. and that won the first place at "Large Scale Visual Recognition Challenge 2015" (ILSVRC2015).

Filter by
Sorted by
Tagged with
0 votes
1 answer
41 views

Why do deep neural networks(not residual nets) suffer from poor performance?

I read through the ResNets paper by Kaiming He et al. They say it is not a vanishing gradient problem but instead it is the inability of a layer to learn the identity function. Why does this problem ...
0 votes
1 answer
52 views

How to handle the size difference of highway network or residual network in cnn?

For highway network, it looks like this: For residual network, it looks like this: Pictures are from What is the name of this neural network architecture with layers that are also connected to non-...
0 votes
0 answers
12 views

Found input variables with inconsistent numbers of samples

I have an issue. the model gave me an error of Found input variables with inconsistent numbers of samples: But I don't understand why ...
  • 1
0 votes
1 answer
53 views

How do I use ResNet for text processing?

I need to implement a deep neural network [residual neural network (ResNet)] that takes some text as an input [length M x N] and then processes it. Now as far as my understanding goes, ResNet is used ...
  • 1
2 votes
0 answers
31 views

Can Inception-ResNet be inverted layer-by-layer?

It has already been shown that by using a normalization layer during training, it is possible to invert a residual network layer-by-layer. I wonder how similar Inception-ResNet is and whether a ...
0 votes
0 answers
26 views

What underlying network is typically meant with ResNET?

When people talk about a ResNet architecture, they are talking about a neural network architecture with skip connections. But what basis network are they typically referring to? Feedforward-networks ...
0 votes
0 answers
44 views

Are ResNets necessary if we use Batch Normalisation?

One of the issues of very deep neural networks is vanishing gradients. This problem was addressed through ResNet by adding skip connections. However, is this actually necessary if we use batch ...
0 votes
0 answers
65 views

In a neural ODE, if $f$ is represented by a neural network, how does it depend on $t$?

In a neural ODE $$\frac{d h(t)}{d t} = f(h(t), t, \theta),$$ we call $f(h, t, \theta)$ the neural vector field. If $f$ is represented by a neural network, how does it depend on $t$? If $t$ is an ...
0 votes
1 answer
28 views

Keep weights of output layer in transfer learning?

I'm seeing conflicting info on what to do with the fully-connected output layer of a pre-trained network when it's used in transfer learning. A previous answer seems to imply that the network is kept ...
0 votes
0 answers
15 views

Loss function when training on binary_accuracy metric

When training a resnet model with metrics set to binary accuracy does the loss function need to be a binary loss function? Now I am using a custom asymmetric loss function (as my data is very ...
  • 1
0 votes
1 answer
175 views

How can I use larger input images when using a pre-trained CNN without resizing?

I have a ResNet18 model trained on the Places365 image dataset, and I'd like to use this pre-trained model to expedite the training needed to identify distressed houses. My dataset is images of size ...
0 votes
1 answer
147 views

Whats wrong with my resnet50 training on CIFAR10 pytorch?

I've been trying to construct resnet50 architecture from scratch using pytorch for classification. After construction I've run training job on CIFAR10 torchvision dataset, in 20 epochs with lr of 0.01 ...
  • 3
0 votes
0 answers
58 views

Exact definition of WRN-d-k (Wide ResNet)

I am a little confused about the WRN-d-k notation from Wide Residual Networks. To quote the paper, In the rest of the paper we use the following notation: WRN-n-k denotes a residual network that has ...
  • 171
0 votes
1 answer
158 views

Retraining ResNet-50 for iris flower classification

I am trying to retrain ResNet-50 for iris flower classification in tensorflow (TensorFlow version: 2.3.0) using the following code ...
  • 11
0 votes
1 answer
59 views

Darknet as a part of Yolo v3

I am pretty new to ML and my question may look strange. Especially the last part of it. 1)As far as I understand Darknet53 is an integral part of Yolo just as Resnet50 is a part of R-CNN Am I right? 2)...
  • 159
0 votes
0 answers
14 views

Why in coordinate regression task Resnet34 performs better than Resnet50?

I'm doing a coordinate regression task for an object. something like this (not exactly this), where the label would be (17,40). So, when I'm using Resnet50 I'm getting poorer results as compared to ...
0 votes
0 answers
38 views

Multiclass image classification - what approach to use and which models to consider?

I'm working on an image classification project and I need to train a multiclass, multilabel classifier. The dataset is large and some of the images are mislabeled (for a given class, some labels are ...
2 votes
2 answers
224 views

How many unique angles of an object do you need in your image training set in order to correctly classify it?

I'm interested in using ResNet-50 to classify images of objects for around 1000 unique classes. I'm wondering if there is any way to estimate how many unique angles I need in my training set to ...
1 vote
1 answer
231 views

How can I use a ResNet as a function approximator for pixel based reinforcement learning?

I'd like to use a residual network to improve learning in image-based reinforcement learning, specifically on Atari Games. My main question is divided into 3 parts. Would it be wise to integrate a ...
2 votes
0 answers
15 views

Why actual mapping is called as unreferenced mapping in this context of residual framework?

Consider the following statements from the research paper titled Deep Residual Learning for Image Recognition by Kaiming He et al. #1: We explicitly reformulate the layers as learning residual ...
  • 3,451
1 vote
2 answers
848 views

Residual Blocks - why do they work?

I've learnt that idea that the residual block was invented to solve the vanishing gradient problem due to the deep layer to layer multiplication. I understand that for example if I have 10 layers, and ...
1 vote
3 answers
196 views

Why do skip layer connections require the same layer sizes?

I know how skip connections work: you add the activations of the previous layer to the activations of a successive layer to stabilize information/gradient flow. My question is, why doesn't it just get ...
  • 296
2 votes
0 answers
42 views

How to implement a (3 + 2)-dimensional convolutional layer where the 2d space is "internal"?

I am trying to train a CNN to learn 5D (kind of) data. The data is structured as follows. It has three spatial dimensions [x, y, z], but it also has two "...
  • 23
1 vote
0 answers
43 views

Does it make sense to apply batch normalization to a batch size of 1?

I am interested in your opinion on the topic if you think that it makes sense to use batch normalization layer in a network that is trained with a batch size of 1. This is a special case as part of an ...
1 vote
1 answer
36 views

Why are weights not initialized with mean=1?

I wonder why weights are initialized with zero-mean. It is one of the reasons, why deep architectures cannot be trained without skip connections. Without the skip connections, the zero initialization ...
  • 31
1 vote
1 answer
239 views

Can residual connections be beneficial when we have a small training dataset?

I have a classification problem, for which an inadequate amount of training data is available. Also, there is no known practical data augmentation approach for this problem (as no unlabelled data is ...
  • 69
1 vote
0 answers
329 views

What are the benefits of Cross Stage Partial Connections over Residual Connections?

Cross Stage Partial Connections (CSPC) try to solve the next problems: Reduce the computations of the model in order to make it more suitable for edge devices. Reduce memory usage. Better ...
1 vote
1 answer
76 views

Regression For Elliptical Curve Public Key Generation Possible?

As part of a learning more about deep learning, I have been experimenting with writing ResNets with Dense layers to do different types of regression. I was interested in trying a harder problem and ...
2 votes
0 answers
441 views

How to use residual learning applied to fully connected networks?

Is there any reason why skip connections would not provide the same benefits to fully connected layers as it does for convolutional? I've read the ResNet paper and it says that the applications should ...
3 votes
1 answer
180 views

Can residual neural networks use other activation functions different from ReLU?

In many diagrams, as seen below, residual neural networks are only depicted with ReLU activation functions, but can residual NNs also use other activation functions, such as the sigmoid, hyperbolic ...
4 votes
1 answer
196 views

If the point of the ResNet skip connection is to let the main path learn the residual relative to identity, why are there convolutional skips?

In the original ResNet paper they talk about using plain identity skip connections when the input and output of a block have the same dimensions. When the input and output have different dimensions ...
6 votes
1 answer
658 views

If vanishing gradients are NOT the problem that ResNets solve, then what is the explanation behind ResNet success?

I often see blog posts or questions on here starting with the premise that ResNets solve the vanishing gradient problem. The original 2015 paper contains the following passage in section 4.1: We ...
3 votes
1 answer
76 views

Do deeper residual networks perform better or worse?

If you have an $18$ layer residual network versus and a $32$ layer residual network, why would the former do better at object detection than the latter, if you have both models are training using the ...
2 votes
2 answers
135 views

What is the name of this neural network architecture with layers that are also connected to non-neighbouring layers?

Consider a feedforward neural network. Suppose you have a layer of inputs, which is feedforward to a hidden layer, and feedforward both the input and hidden layers to an output layer. Is there a name ...
6 votes
1 answer
8k views

Why do ResNets avoid the vanishing gradient problem?

I read that, if we use the sigmoid or hyperbolic tangent activation functions in deep neural networks, we can have some problems with the vanishing of the gradient, and this is visible by the shapes ...
  • 179
1 vote
3 answers
1k views

Is a VGG-based CNN model sometimes better for image classfication than a modern architecture?

I have an image classification task to solve, but based on quite simple/good terms: There are only two classes (either good or not good) The images always show the same kind of piece (either with or ...
  • 155
10 votes
4 answers
6k views

What is the benefit of using identity mapping layers in deep neural networks like ResNet?

As I understand, ResNet has some identity mapping layers, whose task is to create the output as the same as the input of the layer. The ResNet solved the problem of accuracy degrading. But what is the ...
1 vote
1 answer
2k views

Why are there transition layers in DenseNet?

The DenseNet architecture can be summarizde with this figure: Why there are transition layers between each block? In the papers, they justify the use of transition layers as follow : The ...
  • 371
7 votes
1 answer
822 views

Why aren't there neural networks that connect the output of each layer to all next layers?

Why aren't there neural networks that connect the output of each layer to all next layers? For example, the output of layer 1 would be fed to the input of layers 2, 3, 4, etc. Beyond computational ...
5 votes
1 answer
115 views

Are neurons instantly feed forward when input arrives?

Let's say I have a neural network with 5 layers, including the input and output layer. Each layer has 5 nodes. Assume the layers are fully connected, but the 3rd node in the 2nd layer is connected to ...
  • 389
14 votes
2 answers
772 views

Should deep residual networks be viewed as an ensemble of networks?

The question is about the architecture of Deep Residual Networks (ResNets). The model that won the 1-st places at "Large Scale Visual Recognition Challenge 2015" (ILSVRC2015) in all five main tracks: ...