Questions tagged [deep-learning]

For questions related to deep learning, which refers to a subset of machine learning methods based on artificial neural networks (ANNs) with multiple hidden layers. The adjective deep thus refers to the number of layers of the ANNs. The expression deep learning was apparently introduced (although not in the context of machine learning or ANNs) in 1986 by Rina Dechter in the paper "Learning while searching in constraint-satisfaction-problems".

609 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
0
votes
1answer
49 views

How do we make our outputs to have the same size as the true mask?

When we are doing multi-label segmentation tasks, our y_true (the mask) will be (w, h, 3), but, in our model, at the last layer, ...
0
votes
0answers
47 views

Predict probability of user making a conversion

My dear friends, In the past couple of years I read a lot about AI with JS and some libraries like TensorFlow. I have great interest in the subject but never used it on a serious project. However, ...
0
votes
0answers
39 views

Using a neural network in a microcontroller to recognize air-written letters and numbers

Recently I finished a project which combined stm32f030r8 microcontroller and MPU9250 sensor to create a system which would detect orientation on all 3 axis using a combination of accel data, gyro data ...
0
votes
0answers
43 views

CIFAR-10 can't get above 10% Accuracy with MobileNet, VGG16 and ResNet on Keras

I'm trying to train the most popular Models (mobileNet, VGG16, ResNet...) with the CIFAR10-dataset but the accuracy can't get above 9,9%. I want to do that with the completely model (include_top=True) ...
0
votes
2answers
126 views

Practical skills in AI

For the past month I have been studying machine/deep learning, thus I have completed both the machine learning course and the deep learning specialization by Prof. Andrew Ng on coursera. In these ...
0
votes
0answers
28 views

How do I get my DCGAN to generate a number of fake images?

I have a Deep Convolutional Generative Adversarial Network (DCGAN) that trains on the CIFAR dataset. When I finish the training (100k epochs), how can I make my network generate 1000 fake images? I ...
0
votes
2answers
154 views

Why isn't the loss of my neural network reduced after 2500 iterations?

I have developed a basic feedforward neural network from scratch to classify whether image is of cat or not cat. It works fine, but after 2500 iterations, my cost function is not reducing properly. ...
0
votes
0answers
49 views

What are the differences between backbones, frontends, models and architectures in applied deep learning?

Context I'm trying to dive into deep learning for tasks on images, and trying to figure out how to reuse some well-known structures* that have been published, mainly on github. ( *Here, structure can ...
0
votes
0answers
23 views

Different result from k-cross validation model and Train-Validation-Test split model ? (AI fresher question)

I am starting to learn about Neural Network and I have come into one problem that I am still learning how to figure it out. I have a dataset with shape (105,96) (105 samples and 95 first column as ...
0
votes
0answers
22 views

Best ROC threshold for classifier?

Suppose I have a neural network $N$ that produces the output probabilities $[0.3, 0.8]$. Normally, I would specify a threshold of 0.5 for the argmax of the prediction, let's say, second arg > 0.5 ...
0
votes
0answers
30 views

How to feed the LSTM with different length for the latest time step?

I am having a training data set for a time-series dataset like below: ...
0
votes
0answers
22 views

Computational difference between ANN and Pattern Matching

Conceptually, if you had an internal 3d model of all objects in CV you could do a scan matching algorithm. This algorithm would be ridiculously computationally intensive, but it would have a high ...
0
votes
0answers
46 views

Why are YOLO Darknet weights so heavy?

I have been trying to understand how YOLO Darknet works, and for the most part reading the documentation and checking the code helps me understand. But when it comes to the weight file, I can't find ...
0
votes
1answer
73 views

Why is non-linearity desirable in a neural network?

Why is non-linearity desirable in a neural network? I couldn't find satisfactory answers to this question on the web. I typically get answers like "real-world problems require non-linear ...
0
votes
0answers
14 views

Do we train the spatial net and temporal net separately in two stream CNN for action recognition from video frame?

In two-stream CNN paper, they have the following network architecture: We can use a simple average or multiclass linear SVM for the class score fusion. In the case of averaging the outputs of two ...
0
votes
0answers
42 views

How is centralised training and decentralised execution in multi agent reinforcement learning implemented?

In the paper Bayesian Action Decoder for Deep Multi-Agent Reinforcement Learning, it is written We allow centralised training but require decentralised execution, from which follows that the policies ...
0
votes
0answers
21 views

Which model structure is best suited to build a Math OCR (img2latex)? RAM, DRAM, CRNN or Attention OCR?

I am trying to build an OCR which can read the Mathematical equations just like MAthPix and im2markup. im2markup by HarvardNLP seems like a good model but the thing is that it is built using Torch and ...
0
votes
0answers
35 views

Reinforcement Learning Diagnostic: Total reward doesn't converge

I'm implementing DDQN in my toy scenario. During training, I'm surprised to see that the total reward doesn't converge and have a tendency to degrade. What could be the problem? Here's the picture: ...
0
votes
1answer
56 views

What are some good papers or resources for aspect extraction and opinion modelling from video or audio?

I am quite new to deep learning. I just finished the deep learning specialization by Professor Andrew NG and Deep Learning AI. Now, my professor (instructor) has advised me to look into some classic ...
0
votes
0answers
18 views

How to figure out loss weight for label-imbalanced regression problems?

In classification, suppose you have 1 image labeled as cancer and 99 labeled as not cancer, you can just divide the loss weight of "not cancer" by 99. Then you can train the model as this ...
0
votes
0answers
57 views

Normalization of every patch of CNN input separately

What would be the effect of normalizing each input patch going to Convolutional layer separately. Let's say our input is 64 channels of the size 224x224 (like is the case for some hidden layers in ...
0
votes
0answers
15 views

Can attention with 2d position encoding beat capsule on cv tasks?

I have always had doubts about the necessity and intuitive/theoretical justification for capsule network in image classification and more recently nlp tasks. For the former, in order to address the ...
0
votes
0answers
17 views

Separated LSTMs or a global one for cluster of related features

I have an $n$-dimensional time-series to apply LSTM to, $n$ is the number of features for each time point. These features can be clustered according to their concept, for example $n_1, ..., n_4$ are ...
0
votes
0answers
79 views

TD-Leaf struggles at learning chess

I am currently working on implementing Giraffe chess algorithm. Following this paper, I designed a neural network similar to the one proposed by the author which I trained using TD-Leaf(lambda). The ...
0
votes
0answers
43 views

How GAN generator produce integer RGB colored picture?

For traditional neural networks, I know that we can't constraint the output to be strict integers. My question is what technique does GANs use to produce integer outputs, that can be then converted to ...
0
votes
0answers
88 views

Should the network weights converge when training Deep Q networks?

I have two sets of data, one training and one test set. I use the train set to train the deep q network model variant. I also continuously evaluate the agent Q values obtained on the test set every ...
0
votes
0answers
18 views

Identifying and Labeling multiple letters in image

While I attempt to learn AI/ML I have taken on the task to create a Boggle solver. The idea is that a system could take an image of a Boggle arrangement of letters and identify the letters (and the ...
0
votes
0answers
50 views

LSTM for imbalanced panel data

The available tutorials are most focused on time series prediction. I am wondering how shall we prepare the input data when it is an imbalanced data? Here is how data looks like. ...
0
votes
0answers
18 views

How would one modify CycleGAN in order to map a distribution to itself?

CycleGAN can map between two different distributions $X$ and $Y$ with cycle consistency. This is done with generator functions $F: X \mapsto Y$ and $G: Y \mapsto X$, such that $||G(F(x)) - x||_1 \...
0
votes
0answers
135 views

Backward pass of CNN like Resnet: how to manually compute flops during backprop?

I've been trying to figure out how to compute the number of Flops in backward pass of ResNet. For forward pass, it seems straightforward: apply the conv filters to the input for each layer. But how ...
0
votes
0answers
59 views

Reinforcement learning random agent always performing the same few actions

I have a DQN model which has 3 features as inputs (properly normalized) and should output q-values for each of the 100 possible actions. However, prior to any training, I would like to examine the ...
0
votes
0answers
68 views

How does multiple outputs added from intermediate layers of a keras functional model influence its learning behaviour / gradients?

let us assume I have a keras functional model with 2 inputs. My model has two branchs, each branch for each input. The model only uses dense layers. The first input is the data itself (feature vector ...
0
votes
0answers
18 views

Why do code implementations average the loss over a batch instead of finding the expected sample of that batch (using sampling probabilities)

Usually, our training objective over a batch is written in terms of the expected value of a sample in that batch such as $objective = E_{x \sim data} * log(P(x))$ But in the code implementations, ...
0
votes
0answers
28 views

How to stabilize the training of a Conv-Siamese Neural Network if the results after different trainings vary relatively strongly?

I am training a neural network using MSE and ADAM optimizer. More precisely, a siamese architecture with a convolutional encoder and euclidean distance on top. I am using MSE because I have different ...
0
votes
0answers
20 views

Deep Network with constraint or auxiliary features

The target of my current neural network is to predict a label. The dataset contains some features, there is a label $y_i$ in transaction $i$, indicating its classification. There is one feature $f^{i}...
0
votes
0answers
40 views

Are there any papers on this alternate neural net training approach?

I developed a custom callback for Keras. Initially, it monitors training accuracy. If on a given epoch the accuracy is below that of the previous epoch it lowers the learning rate by a factor. If for ...
0
votes
0answers
32 views

Can neural networks always be assembled like Lego blocks?

BACKGROUND Consider a supervised problem which is based on two scalar features (1) and (2) as well as a third, "time-dependent", feature consisting of a sequence of five values (3)-(7). For ...
0
votes
0answers
16 views

Can I use pertained ASR model for SER purpose?

Can I use pertained ASR(Automatic speech recognition) models like deep speech, jasper, to name a few for speech emotion recognition purposes? I was thinking by using the features which are extracted ...
0
votes
0answers
13 views

One model to output the whole vector vs different model to output each vector element

Let's say I'm using a neural net to solve some problem where the output of the net is a vector of some size. What are the advantages (and disadvantages if there are any) of training a single net to ...
0
votes
0answers
34 views

What is the degree of linearity in the error propagated by Gradient Descent?

Neural Network is trained to learn a non-linear function, the more layers it has, the more is the quality of the prediction and the ability to match the real-world function correctly (lets leave aside ...
0
votes
0answers
48 views

Where can I get exercises and problems to implement machine learning models and algorithms?

I'm an intermediate machine learning student and want to get more detailed and specific practical intuition about artificial intelligence. I have made a couple of searches over the well-observed ...
0
votes
0answers
16 views

How can I perform object detection by cutting the image into many pieces each containing one object?

Our task is to do a special object detection: In the traditional case, the neural network will output some rectangle bounding boxes. But in our case, the network should output many nearly-vertical ...
0
votes
0answers
66 views

Intuition behind single-shot object detection

Is there a good way to understand how single-shot object detection works? The most basic way to do detection is use a sliding-window detector and look at the output of the NN to detect if a class is ...
0
votes
0answers
113 views

How to calculate covariance matrix of the mini-batch in the k-th layer using Python?

I am a beginner in Python. I want to calculate the covariance matrix of a mini-batch in a given hidden layer.
0
votes
0answers
144 views

Can a neural net learn to read?

I am a student of last year of computer engineering and lately I have been very interested in AI. Fields such as ML and DL seem very disruptive to me. A few months ago I saw an interview of Bill ...
0
votes
0answers
54 views

Pooling vs Subsampling: Multiple Definitions?

I have seen people using pooling and subsampling synonymously. I have also seen people use them as different processes. I am not sure though if I have correctly inferred what they mean, when they use ...
0
votes
0answers
155 views

How do I do further (domain-specific) pre-training with Google BERT in preparation for subsequent fine-tuning?

How do I do further (domain-specific) pre-training with Google BERT in preparation for subsequent fine-tuning? Another way to say this is: can you create a checkpoint file created from the final ...
0
votes
0answers
29 views

Deep Generative Networks Probability of “Success”

I have built various "successful" GANs or VAEs that can generate realistic images reliably, but in either case the generative step is sampling a latent feature vector from some distribution and ...
0
votes
0answers
19 views

Consecutive frames can be discarded when training an SSD/YOLO?

Let's say I have a number of videos, and I want to train an SSD/YOLO (or FRCNN) to detect objects. In the case of a large amount of videos, there will be a lot of frames extracted and transferred to ...

1
9 10 11
12
13