Hottest Questions This Month

2

votes

1

answer

1k

views

Should I continue training if the neural network attains 100% training accuracy?

modified Jul 6 at 14:23 spiridon_the_sun_rotator 583

1

vote

3

answers

113

views

Can people use neural networks without providing the set of training data?

neural-networks machine-learning reinforcement-learning training-datasets

modified Jul 6 at 12:50 nbro♦ 28k

3

votes

2

answers

191

views

How can we find the value function by solving a system of linear equations?

reinforcement-learning value-functions bellman-equations linear-algebra

modified Jul 7 at 20:44 Neil Slater 19.4k

2

votes

1

answer

222

views

Is the dropout technique specific only to neural networks?

neural-networks regularization decision-tree dropout

answered Jul 18 at 13:07 harwiltz 911

1

vote

3

answers

161

views

Why is AI Super Resolution Reconstruction more than just guessing?

computer-vision papers image-processing image-super-resolution tecogan

answered Jul 2 at 19:37 generic_stackoverflow_user 1

3

votes

2

answers

77

views

Is there any difference between affine transformation and linear transformation?

neural-networks terminology pytorch linear-algebra transformation

modified Jul 19 at 10:43 y-prudent 21

1

vote

2

answers

69

views

Is there any model that is probabilistic but not statistical?

machine-learning terminology statistical-ai

modified Jul 1 at 13:52 nbro♦ 28k

2

votes

1

answer

106

views

Not able to find a good fit for a simple function with neural networks

deep-learning

modified 23 hours ago crinix 236

1

vote

1

answer

75

views

How much C++ is needed for research in machine learning and artificial intelligence?

academia programming-languages software-evaluation education

answered Jul 9 at 20:41 nbro♦ 28k

0

votes

1

answer

58

views

Which policy do I need to use in updating Q function?

reinforcement-learning q-learning terminology off-policy-methods exploration-strategies

modified Jul 1 at 1:25 satya 187

1

vote

1

answer

67

views

What does it mean by overfitting the test set?

machine-learning terminology overfitting test-datasets

modified Jul 7 at 9:30 nbro♦ 28k

0

votes

1

answer

75

views

How can I approach this problem of producing a 3-bit binary string given a sequence of letters?

classification

modified Jul 15 at 13:00 user366312 103

3

votes

1

answer

44

views

Can some of the weights be fixed during the training of a neural network?

machine-learning convolutional-neural-networks training backpropagation weights

modified Jul 6 at 15:43 nbro♦ 28k

1

vote

1

answer

53

views

Which policy has to be followed by a player while construction of its own Q-table?

reinforcement-learning q-learning off-policy-methods exploration-exploitation-tradeoff exploration-strategies

modified Jul 3 at 2:27 nbro♦ 28k

2

votes

2

answers

64

views

Why adversarial images are not the mainstream for captchas?

computer-vision adversarial-ml captcha

modified Jul 4 at 11:12 Aray Karjauv 476

1

vote

1

answer

62

views

In addition to the reward function, which other functions do I need to implement Q-learning?

reinforcement-learning q-learning markov-decision-process reward-functions exploration-strategies

modified Jul 1 at 10:00 satya 187

1

vote

1

answer

48

views

Why do I get bad results no matter my neural network function approximator for parametrized Q-learning implementation for Contextual Bandits?

reinforcement-learning q-learning dqn multi-armed-bandit contextual-bandit

modified Jul 9 at 9:04 Neil Slater 19.4k

1

vote

1

answer

52

views

Does regularization just mean using an augmented loss function?

terminology resource-request regularization

answered 4 hours ago crinix 236

2

votes

1

answer

49

views

Why is tanh a “smoothly” differentiable function?

neural-networks terminology math activation-function tanh

answered Jul 8 at 13:38 htl 674

1

vote

1

answer

75

views

What math should I learn before and while using and applying deep learning?

deep-learning math

modified Jun 28 at 13:06 nbro♦ 28k

1

vote

1

answer

35

views

What exactly is an XPU?

terminology pytorch resource-request hardware

answered Jul 23 at 8:20 JVGD 718

2

votes

2

answers

34

views

How to determine the embedding size?

deep-learning hyperparameter-optimization hyper-parameters embeddings

modified Jul 11 at 13:06 nbro♦ 28k

0

votes

0

answers

67

views

Will AI replace software developers?

open-ai software-development gpt-3

modified Jul 1 at 11:10 nbro♦ 28k

1

vote

1

answer

52

views

How much money is spent training neural networks each year by companies such as Google and Facebook?

neural-networks deep-learning deep-neural-networks google

answered Jul 6 at 12:14 Recessive 1,196

1

vote

1

answer

39

views

What does it mean by “zeros the networks parameters gradients” in the context of training a neural network?

neural-networks training pytorch gradient

answered Jul 18 at 5:09 spiridon_the_sun_rotator 583

0

votes

1

answer

50

views

What is the best way to stay up to date with state of the art machine learning? [duplicate]

machine-learning deep-learning research academia state-of-the-art

modified Jul 8 at 10:59 Saurav Maheshkar 624

5

votes

0

answers

46

views

Proof that there always exists a dominating policy in an MDP

reinforcement-learning markov-decision-process proofs policies

modified Jul 16 at 17:52 nbro♦ 28k

1

vote

1

answer

44

views

What are (all) the differences between a neuron and a perceptron?

comparison perceptron neurons

modified Jul 15 at 15:44 nbro♦ 28k

0

votes

1

answer

44

views

The model learns well, but the validation decreases over time [closed]

reinforcement-learning natural-language-processing

answered Jul 19 at 12:05 Mohammed Abed 26

0

votes

1

answer

34

views

Why would the “improvement” be the result of random initialization, and so why should we use multiple runs?

transfer-learning weights-initialization multiclass-classification uncertainty-quantification k-fold-cv

modified Jul 5 at 12:54 nbro♦ 28k

0

votes

1

answer

34

views

What are the labels of EMNIST?

neural-networks datasets pytorch mnist

modified Jul 15 at 21:38 Avatron 136

1

vote

1

answer

42

views

Why doesn't the LSTM model improve the time-series forecasting significantly with respect to the MLP model?

deep-learning long-short-term-memory time-series multilayer-perceptron forecasting

answered Jul 17 at 0:23 crinix 236

1

vote

2

answers

32

views

Train Validation Test Splitting After or Before Data Augmentation?

training datasets data-augmentation

answered Jul 22 at 7:56 vlthr 36

0

votes

1

answer

52

views

Can the developers of the libraries like TensorFlow have access to my dataset when I develop the model on my PC?

tensorflow keras datasets security scikit-learn

modified Jul 2 at 6:45 Sai Prashanth 1

1

vote

1

answer

40

views

Why is the input layer of a neural network usually not counted?

neural-networks hyper-parameters input-layer

modified Jul 8 at 11:57 nbro♦ 28k

1

vote

1

answer

24

views

Book(s) on generative models

generative-model resource-request

answered Jul 4 at 13:21 spiridon_the_sun_rotator 583

0

votes

1

answer

24

views

Is it possible that the fine-tuned pre-trained model performs worse than the original pre-trained model?

tensorflow object-detection transfer-learning pretrained-models fine-tuning

modified Jul 14 at 11:29 nbro♦ 28k

2

votes

1

answer

39

views

Are there any advantages of the local attention against convolutions?

convolutional-neural-networks attention

answered Jul 10 at 21:45 TimD1 131

1

vote

1

answer

38

views

How is the convolution operation connected to neural networks?

neural-networks convolutional-neural-networks convolution

modified Jul 17 at 17:05 nbro♦ 28k

1

vote

1

answer

38

views

In the cross-entropy method, should I select state-action pairs by their immediate reward or by the episode reward?

reinforcement-learning rewards monte-carlo-methods return cross-entropy-method

modified Jul 22 at 11:42 nbro♦ 28k

-1

votes

1

answer

29

views

how produce image of face working with AI [closed]

image-processing

answered Jul 23 at 16:14 Aray Karjauv 476

1

vote

1

answer

36

views

What exactly is the population in the problem of finding the best path in a network of nodes using genetic algorithms?

genetic-algorithms genetic-programming path-finding

modified Jul 5 at 15:53 shadi 11

0

votes

2

answers

46

views

How to design a neural network with arbitrary input and output length?

neural-networks machine-learning deep-learning recurrent-neural-networks pretrained-models

modified Jul 13 at 10:57 nbro♦ 28k

2

votes

1

answer

27

views

Why adding a baseline doesn't affect the policy gradient?

policy-gradients proofs open-ai reinforce

modified Jul 11 at 13:04 nbro♦ 28k

1

vote

2

answers

34

views

Example of lemma having multiple boldface forms

natural-language-processing books

modified Jun 28 at 17:52 Golden Lion 119

1

vote

1

answer

34

views

In Computer Vision, what is the difference between a transformer and attention?

computer-vision comparison transformer attention

answered yesterday nbro♦ 28k

0

votes

1

answer

33

views

Where does batch normalization layers present in a neural network?

neural-networks architecture batch-normalization

modified Jul 22 at 3:32 hanugm 1,172

0

votes

0

answers

42

views

Does randomly adding hand-engineered features increase the CNN's sample efficiency/performance?

convolutional-neural-networks computer-vision feature-engineering sample-efficiency

modified Jul 14 at 20:48 nbro♦ 28k

Top Questions

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