Hottest Questions Today

0

votes

1

answer

17

views

Is it a good practice to pad signal before feature extraction?

answered 2 hours ago Kostya 759

0

votes

0

answers

18

views

Saving Computational Time with DNN Pruning

deep-learning time-complexity computational-complexity

asked 5 hours ago Stefano Barone 183

1

vote

1

answer

23

views

Off policy Temporal Difference learning control using Q values

machine-learning reinforcement-learning temporal-difference-methods importance-sampling

modified 2 hours ago Neil Slater 18k

0

votes

0

answers

12

views

issue with gradient notation in stochastic value gradient paper

backpropagation papers

asked 6 hours ago quest ions 322

0

votes

0

answers

12

views

How to use Intel's pretrained pose estimation models and decode the output

pose-estimation

asked 8 hours ago Yixuan Wang 43

-1

votes

0

answers

22

views

AI Researchers needed to interview [closed]

research

asked 12 hours ago Marcus Casillas 1

0

votes

0

answers

18

views

Keras layer to keep the maximum of time series [closed]

tensorflow keras long-short-term-memory time-series

modified 11 hours ago Saurav Maheshkar 445

-1

votes

0

answers

21

views

Are there AI black projects? [closed]

neural-networks deep-learning

modified 15 hours ago Edison 99

0

votes

0

answers

16

views

LXMERT huggingface [closed]

deep-learning natural-language-processing computer-vision transformer attention

asked 17 hours ago Mariam 1

0

votes

1

answer

38

views

On what basis is MATLAB “inflexible” to perform ML/AI research on it?

research matlab

modified yesterday nbro♦ 26.4k

1

vote

1

answer

30

views

Unable to 'learn' a rotational angle by parametrising the angle as a neural network layer

deep-neural-networks

modified 18 hours ago damith219 13

2

votes

1

answer

39

views

Are neural networks invertible?

neural-networks

answered yesterday htl 602

-1

votes

0

answers

13

views

Terrain slopes on HALF-CHEETAH pybullet [closed]

deep-rl gym proximal-policy-optimization continuous-action-spaces pybullet

asked yesterday Shibhansh Dohare 1

0

votes

0

answers

21

views

Is there a (deep) RL algorithm that can map combinations of multiple states into corresponding combinations of multiple actions?

neural-networks reinforcement-learning deep-rl robotics algorithm-request

modified yesterday nbro♦ 26.4k

3

votes

1

answer

194

views

What does “statistical efficiency” mean in this context?

convolutional-neural-networks terminology books

modified 2 days ago nbro♦ 26.4k

0

votes

0

answers

13

views

Can transformers be used to improve regression?

neural-networks transformer regression function-approximation

asked yesterday cyberface 19

0

votes

1

answer

33

views

CNN based model poor result

convolutional-neural-networks pytorch

modified yesterday SmarArror 26

0

votes

1

answer

18

views

How to train my model using transfer learning on inception_v3 pre-trained model?

tensorflow transfer-learning pretrained-models inception

modified yesterday Saurav Maheshkar 445

0

votes

1

answer

54

views

Why is loss displayed as a parabola in mean squared error with gradient descent?

loss-functions gradient-descent

modified yesterday SmarArror 26

-2

votes

0

answers

24

views

RL with tabular state instead of image

reinforcement-learning

asked yesterday jchaykow 169

2

votes

1

answer

22

views

Why object detection algorithms are poor in optical character recognition?

object-detection object-recognition optical-character-recognition ctc-loss

answered yesterday Mike NZ 201

5

votes

2

answers

49

views

What does “semantic gap” mean?

machine-learning terminology papers transfer-learning

answered yesterday DukeZhou♦ 6,030

1

vote

1

answer

31

views

How to design fitness function for multiple objectives?

neural-networks genetic-algorithms fitness-functions

modified yesterday Rasmus 13

1

vote

1

answer

27

views

How to handle class imbalancing when the actual data are that way

classification training data-preprocessing supervised-learning imbalanced-datasets

answered 2 days ago Mike NZ 201

0

votes

0

answers

11

views

What's new in LaBSE v2?

tensorflow bert

asked 2 days ago Brainless 101

-1

votes

0

answers

17

views

Weather detection by images [closed]

image-recognition training-datasets

asked 2 days ago Francesco Pagani 1

-1

votes

0

answers

14

views

How can I implement a sigmoid function composed from three distinct phases? [closed]

sigmoid utility-based

asked 2 days ago stevGates 99

0

votes

1

answer

28

views

How to source training data in ML for information security?

training datasets resource-request security training-datasets

modified yesterday Saurav Maheshkar 445

0

votes

0

answers

14

views

How to create a fitness function suitable for a recommender system problem?

genetic-algorithms evolutionary-algorithms recommender-system fitness-functions

modified 2 days ago nbro♦ 26.4k

-1

votes

1

answer

22

views

Using states (features) and actions from a heuristic model to estimate the value function of a reinforcement learning agent [closed]

reinforcement-learning deep-rl dqn control-problem policy-evaluation

answered 2 days ago Kristof 26

0

votes

0

answers

13

views

Training labels: integers or vectors?

tensorflow q-learning dqn loss-functions

asked 2 days ago Kricket 197

0

votes

1

answer

43

views

Why can a neural network use more than one activation function?

math activation-function function-approximation universal-approximation-theorems

modified yesterday Saurav Maheshkar 445

1

vote

0

answers

23

views

What is a Hebbian linear classifier?

machine-learning deep-learning papers hebbian-learning

modified Apr 18 at 13:27 Oliver Mason 4,159

0

votes

2

answers

55

views

In variational autoencoders, why do people use MSE for the loss?

loss-functions autoencoders variational-autoencoder mean-squared-error evidence-lower-bound

answered Apr 18 at 9:56 IttayD 119

0

votes

0

answers

48

views

Comparing heuristics in A* search and rescue operation

research a-star path-finding heuristic-functions

modified Apr 17 at 5:07 Faizy 551

0

votes

0

answers

22

views

Is there any way to determine/estimate the number of rounds for the whole Federated Learning process?

machine-learning hyperparameter-optimization hyper-parameters federated-learning

modified Apr 17 at 21:44 nbro♦ 26.4k

5

votes

1

answer

88

views

Does the policy iteration convergence hold for finite-horizon MDP?

reinforcement-learning markov-decision-process convergence policy-iteration

modified Apr 16 at 22:53 Kostya 759

0

votes

1

answer

50

views

+50

Does a differential evolution algorithm mutate its population during a generation?

search evolutionary-algorithms meta-heuristics pseudocode differential-evolution

answered Apr 17 at 18:33 Kostya 759

1

vote

0

answers

29

views

Could the inputs of the mean squared-error loss function be transformed to allow larger learning rates?

neural-networks loss-functions math gradient-descent loss

modified Apr 17 at 12:14 Saurav Maheshkar 445

4

votes

2

answers

47

views

What are the pros and cons of using sigmoid or softmax approach when dealing with 2 classes?

neural-networks machine-learning deep-learning activation-function

answered Apr 16 at 1:31 nxglogic 111

-1

votes

1

answer

30

views

How can the gradient of the weight be calculated in the viewpoint of matrix calculus?

gradient-descent linear-algebra calculus

answered 2 days ago 최선웅 99

0

votes

0

answers

22

views

Identifying if a model is over or under-fitting via graphs

neural-networks machine-learning overfitting performance underfitting

asked Apr 17 at 2:44 jr123456jr987654321 91

0

votes

1

answer

36

views

Decision boundary figure in Least square GAN paper

deep-learning generative-adversarial-networks generative-model

answered Apr 17 at 22:16 Kostya 759

0

votes

0

answers

28

views

What are the math theorems regarding the Multilayer Perceptron?

neural-networks reference-request math multilayer-perceptron perceptron

modified Apr 16 at 16:52 Verónica Rmz. 121

0

votes

1

answer

27

views

Are there architectures to generate pictures from four labels? (VAEs, GANs)

generative-adversarial-networks variational-autoencoder

answered Apr 16 at 13:42 Aray Karjauv 336

0

votes

1

answer

21

views

What's the difference between a 1d tensor and a 2d tensor with 1 dimension?

machine-learning tensorflow math linear-algebra

modified Apr 16 at 13:32 nbro♦ 26.4k

1

vote

2

answers

59

views

In variational autoencoders, what does p(x|z) mean?

papers variational-autoencoder notation

answered Apr 15 at 19:35 David Ireland 3,559

2

votes

0

answers

26

views

Does Yann LeCun consider k-means self-supervised learning?

terminology unsupervised-learning self-supervised-learning

modified Apr 16 at 10:10 Pieter 121

Top Questions

Looking for more? Browse the complete list of questions, or popular tags. Help us answer unanswered questions.