We’re rewarding the question askers & reputations are being recalculated! Read more.

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".

Filter by
Sorted by
Tagged with
0
votes
0answers
10 views

Neural Network for dimension estimation of primitive shapes?

Is it possible to train a Convolutional Network to predict the dimensions of primitive objects such as (spheres, cylinders, cuboids .. ) from point clouds ? Given to the Network will be the point ...
2
votes
1answer
45 views

What are causative and exploratory attacks in Adversarial Machine Learning?

I've been researching Adversarial Machine Learning and I know that causative attacks are when an attacker manipulates training data. An exploratory attack is when the attacker wants to find out about ...
1
vote
0answers
9 views

Where to find implementation of preprocess_input in Keras / Tensorflow 2.0?

When you use a pretrained network such as VGG, ResNet, etc. it used to be that the they all had corresponding preprocess_input functions directly implemented in the functions that were being called/...
2
votes
1answer
29 views

Pose estimation using CNNs on Point clouds

In the case of single shot detection of point clouds, that is the point cloud of an object is taken only from one camera view without any registration. Can a Convolutional Network estimate the 6d pose ...
0
votes
1answer
45 views

Do algorithms like OpenAI's “think up strategies”?

I was discussing with a friend whether current AI does anything remotely similar to 'thinking' and he argued that AIs that play games must think up strategies. While thinking may not be precisely ...
2
votes
1answer
50 views

Is there anything theoretically revolutionary about Deep Neural Network?

In recent years we have seen quite a lot of impressive display of Deep Neural Network (DNN), as demonstrated most famously by AlphaGo and its cousin programs. But if I understand correctly, deep ...
2
votes
0answers
40 views

Which deep neural networks are appropriate for the detection of bombs?

This is a follow-up question from my previous post here about explosion detection. I gathered a dataset of explosions. As I'm new to Deep Learning in Keras, I'm trying to see what architecture best ...
2
votes
1answer
42 views

AI natural voice generator

I want to create a solution, which clones my voice. I tried my commercial solutions or implementation of Tacotron. Unfortunately, results not sound natural, generated voice sounds like a robot. ...
2
votes
0answers
57 views

Is there a dataset for the detection of bomb explosions?

I would like to train a deep neural network to recognize bomb explosions. I was wondering if there is an open visual dataset for bomb explosion? Alternatively, if you know a good deep architecture or ...
2
votes
1answer
32 views

How exactly does adversarial training help in handling mode-collapse in generative networks?

Of my understanding mode-collapse is when there happen to be multiple classes in the dataset and the generative network converges to only one of these classes and generates images only within this ...
3
votes
1answer
33 views

Why is Jensen-Shannon divergence preferred over Kullback-Leibler divergence in measuring the performance of a generative network?

I have read articles on how Jensen-Shannon divergence is preferred over Kullback-Leibler in measuring how good a distribution mapping is learned in a generative network because of the fact that JS-...
2
votes
1answer
46 views

Is it possible to create a decompiler using AI?

I am trying to decode a compiled file to source code and I am failing. I want to know whether an AI based decompilation is possible for a compiled files? Is it possible to create a decompiler using a ...
0
votes
0answers
17 views

How can I use deep reinforcement learning for vehicle rerouting in SUMO?

I want to use deep reinforcement learning for vehicle rerouting in SUMO, but I don't know how to start training the model. I've already created road network and vehicle routing in SUMO-XML files (...
1
vote
0answers
27 views

AlphaGo Zero: Does the policy head give a probability for every possible move?

If I understood correctly, the AlphaGo Zero network returns two values: a vector of logit probabilities p and a value v. My question is: in this vector that it is outputted, do we have a probability ...
4
votes
1answer
225 views

How to create an AI to solve a word search?

This at first sounds ridiculous. Of course there is an easy way to write a program to solve a wordsearch. But what I would like to do is write a program that solves a word-search like a human. That ...
0
votes
0answers
36 views

Training with many CPU cores doesn't improve performance

I ran my job on a computing cluster: first with 4 cores, then with 32 cores (2 nodes). But the training time is pretty much exactly the same for both of them: ~67 seconds per step. I am trying to ...
0
votes
0answers
7 views

How to set loss weight to zero for an output dimension in keras?

Suppose I am training a model to detect facial keypoints that allow occlusions to be present. The input is an image of a face, and the model has to predict the x,y coordinate of both eyes and mouth. ...
1
vote
0answers
13 views

How we can keras.Imagedatagenerator from csv file? [on hold]

How we can use Imagedatagenerator to produce images if we have a folder with a name "train" which contain all the images of every classification( folder contains mixed images). And we have one csv ...
1
vote
0answers
19 views

How to implement CNN with variable number of images in tensorflow or keras?

Suppose I have a problem where I want to classify the color of LEDs seen in the image. I can use OpenCV to pinpoint the exact location of these LEDs but I do not know their color for sure because the ...
1
vote
0answers
11 views

Deep audio fingerprinting for word search

Simply speaking, I'm trying to somehow search an audio clip for a list of words, and if found, I mark the time stamps. My use-case is profanity check with a list of pre-defined profane words. Is ...
1
vote
1answer
34 views

Detailed explaination of Facenet Model for face recogniton?

Can some one explain how Facenet model works in detail and simple words .
1
vote
2answers
41 views

Why can't LSTMs tell a long story?

There is a recent trend in people using LSTMs to write novels. I haven’t attempted this myself. From what I’m hearing, they can tell a story, but it seems they lose the context of the story rather ...
3
votes
1answer
66 views

Are deep learning models more prone to overfitting than machine learning ones?

In my opinion, deep learning algorithms and models (that is, multi-layer neural networks) are more sensitive to overfitting than machine learning algorithms and models (such as the SVM, random forest, ...
0
votes
0answers
15 views

I am building a SGAN for image classfication. It always predicts the output as class1, even if image from training set is given [closed]

I am building a semi-supervised GAN for image classification to classify the images into two classes, so that I can later use it to classify cyberbully. When I load the model and use it for prediction,...
2
votes
0answers
26 views

Are embeddings in multi-lingual language models comparable across languages?

Facebook has just pushed out a bigger version of their multi-lingual language model XLM, called XLM-R. My question is: do these kind of multi-lingual models imply, or even ensure, that their ...
1
vote
0answers
31 views

How is gradient being calculated in Andrej Karpathy's pong code?

I was going through the code by Andrej Karpathy on reinforcement learning using a policy gradient. I have some questions from the code. Where is the logarithm of the probability being calculated? ...
0
votes
0answers
27 views

How could I locate certain words or numbers in a financial statement?

I would like to code a script that could locate a specific word or number in a financial statement. Financial statements roughly contain the same information, they are however not identical and ...
3
votes
1answer
33 views

Can neural networks deal with unbounded numbers as inputs?

I want to train an ANN. The problem is that the input features are completely unbounded (There are no boundaries as maximum and minimum for them). For example, the following input vectors $(42, ...
0
votes
0answers
27 views

Fully-Configured Deep Learning VM in Python (VirtualBox or VMware)

Often when we start working on any new technology, the most common challenge that we face is "How can I not waste time setting up my environment but directly working on the technology?" If you aren’t ...
2
votes
1answer
69 views

Why is this ResNet50 misclassifying objects?

I'm new to Deep Learning, and I have some conceptual problems. I followed a simple tutorial here, and trained a model in Keras to do image classification on 10 classes of logos. I prepared 10 classes ...
1
vote
2answers
39 views

TensorFlow 2.0 - Normalizing input to DNN (on structured data)

I have a structured dataset of around 100 gigs, and I am using DNN for classification in TF 2.0. Because of this huge dataset, I cannot load entire data in memory for training. So, I'll be reading ...
2
votes
1answer
24 views

Is batch normalization not suitable for non-gaussian input?

I generate some non-Gaussian data, and use two kinds of DNN models, one with BN and the other without BN. I find that the model DNN with BN can't predict well. The codes is shown as follow: <...
1
vote
1answer
22 views

Dataset for floating objects detection

I am looking for dataset that which one I could train model to detect people / boats / surfboards etc. from drone view. Have anyone saw dataset that could be useful for this purpose? I have some ...
7
votes
2answers
574 views

Why use a recurrent neural network over a feedforward neural network for sequence prediction?

If recurrent neural networks (RNNs) are used to capture prior information, couldn't the same thing be achieved by a feedforward neural network (FFNN) or multi-layer perceptron (MLP) where the inputs ...
2
votes
1answer
15 views

Should I use my redundant feature as an auxiliary output or as another input feature?

For example, given a face image, and you want to predict the gender. You also have age information for each person, should you feed the age information as input or should you use it as auxiliary ...
2
votes
0answers
20 views

Why are researchers focused on deep learning based stereo depth/disparity methods instead of non deep learning ones?

In recent years if you are working on stereo depth/disparity algorithms, it seems like you will only ever get your paper accepted to CVPR/ICCV/ECCV if there's some deep learning involved in it. A lot ...
0
votes
1answer
24 views

How can we print weights per iteration in a simple feed forward MLP for an specific class?

im working on a project in which I have to make a multi-layer perceptron with two hidden layers with 3 nodes in each. The target value in my data contains 8 unique values/classes. One of the tasks ...
1
vote
1answer
5 views

Positioning of batch normalization layer when converting strided convolution to convolution + blurpool

I'm trying to replace the strided convolutions of Keras' MobileNet implementation with the ConvBlurPool operation as defined in the Making Convolutional Networks ...
2
votes
0answers
22 views

How am I getting same results 30 times faster than in original HER paper?

I am reproducing the results from Hindsight Experience Replay by Andrychowicz et. al. In the original paper they present the results below, where the agent is trained for 200 epochs. 200 epochs * 800 ...
1
vote
0answers
8 views

Colour palettes for COCO dataset

Is there any link or explanation on how to create Color palettes for coco dataset for segmentation. I see there is a link for cityscapes dataset https://gluon-cv.mxnet.io/_modules/gluoncv/utils/viz/...
3
votes
2answers
49 views

Why they use KL divergence in Natural gradient?

Natural gradient aims to do a steepest descent on the "function" space, a manifold that is independent from how the function is parameterized. It argues that the steepest descent on this function ...
1
vote
1answer
76 views

How to Extract Information from the png Image

Extract Information from the png Image I'm trying to extract some particular information from the image(png). I tried to extract the text using the below code ...
2
votes
0answers
28 views

What are the differences between Yolo v1 and CenterNet?

I recently read a new paper (late 2019) about a one-shot object detector called CenterNet. Apart from this, I'm using Yolo (V3) one-shot detector, and what surprised me is the close similarity between ...
2
votes
1answer
34 views

when should I create a custom loss function?

Hi I'm using neural network to solve a multi regression problem. I'm trying to predict continuous values, to be more specific I'm making a tracking algorithm to track the position of an Object, I'm ...
1
vote
1answer
46 views

When should I use a linear activation instead of ReLU?

I have read this post: How to choose an activation function?. There is enough literature about activation functions, but when should I use a linear activation instead of ReLU? What does the author ...
2
votes
0answers
30 views

What is the underlying model of IBM Watson Assistant and Microsoft LUIS?

As I stated in my question, I would like to know the underlying pipeline and machine learning models that are used to classify intents and identify entities in IBM Watson Assistant and Microsoft LUIS ...
2
votes
0answers
34 views

Is there a deep learning-based architecture for digit localisation?

I'm new to object detectors and segmentation. I want to localize digits on a plate as fast as possible. All images of the dataset are normalized to $300 \times 60$. There are different approaches to ...
2
votes
1answer
66 views

What would be the most effective self-learning algorithm for a 7 player social deduction game?

There's this 7 player social deduction game called Secret Hitler, and I have been trying to find a self-learning AI algorithm to learn how to play this game for a while. Basically, four players are ...
1
vote
0answers
28 views

Reinforcement learning number of episodes per epoch not matching with paper

I am trying to reproduce results presented in this paper. On page 4, the authors state: ... we train for 50 epochs (one epoch consists of 19*2*50 = 1900 full episodes), which amounts to a total ...
1
vote
3answers
80 views

How can neural networks be used to generate rather than classify?

In my experience with Neural Nets, I have only used them to take input vectors and return binary output. But, here in a video, https://youtu.be/ajGgd9Ld-Wc?t=214, Kai Fu Lee, renowned AI Expert shows ...