Hottest Questions This Month

8

votes

3

answers

1k

views

What are the differences between an agent and a model?

answered Feb 9 at 22:43 Peteris 813

4

votes

1

answer

1k

views

Is there a machine learning model that can be trained with labels that only say how “right” or “wrong” it was?

machine-learning reinforcement-learning tensorflow deep-rl models

modified Feb 12 at 16:10 nbro♦ 25.4k

4

votes

2

answers

364

views

Can CNNs be made robust to tricks where small changes cause misclassification?

convolutional-neural-networks reference-request terminology adversarial-ml

modified Feb 18 at 20:42 Dennis Soemers♦ 8,498

5

votes

3

answers

456

views

In Q-learning, wouldn't it be better to simply iterate through all possible states?

reinforcement-learning q-learning exploration-exploitation-tradeoff exploration-strategies

modified Feb 7 at 16:49 nbro♦ 25.4k

6

votes

2

answers

276

views

Is there an ideal range of learning rate which always gives a good result almost in all problems?

neural-networks deep-learning reference-request gradient-descent learning-rate

modified Feb 23 at 10:23 nbro♦ 25.4k

5

votes

2

answers

177

views

Why is “-0.5 * torch.sum(1 + sigma - mu.pow(2) - sigma.exp())” in Pytorch equivalent to the KL?

proofs variational-autoencoder kl-divergence

modified Feb 18 at 3:37 Sycorax 206

2

votes

1

answer

267

views

Is there any difference between ConvNet and CNN?

convolutional-neural-networks terminology

modified Feb 16 at 16:07 raghu 116

2

votes

1

answer

267

views

Can ML/DL solve my classification problem?

machine-learning deep-learning classification

modified Feb 4 at 21:06 Glorfindel 163

5

votes

2

answers

137

views

What is the relation between the context in contextual bandits and the state in reinforcement learning?

reinforcement-learning terminology function-approximation state-spaces contextual-bandit

modified Feb 11 at 17:01 Neil Slater 17.3k

4

votes

1

answer

120

views

Why do feedforward neural networks require the inputs to be of a fixed size, while RNNs can process variable-size inputs?

neural-networks recurrent-neural-networks feedforward-neural-network multilayer-perceptron

modified Feb 5 at 11:26 David Ireland 3,149

4

votes

1

answer

161

views

Why does Q-learning converge under 100% exploration rate?

reinforcement-learning q-learning convergence epsilon-greedy-policy exploration-strategies

modified Feb 21 at 16:53 Community♦ 1

4

votes

1

answer

142

views

Is it practical to train AlphaZero or MuZero (for indie games) on a personal computer?

training game-ai alphazero muzero

modified Feb 15 at 22:35 nbro♦ 25.4k

4

votes

1

answer

141

views

Will there be some promising techniques that can make AI greener and affordable in the future?

machine-learning deep-learning ethics green-ai red-ai

modified Feb 24 at 15:01 Lerner Zhang 571

5

votes

1

answer

94

views

How to improve the reward signal when the rewards are sparse?

reinforcement-learning reward-functions sparse-rewards delayed-rewards potential-reward-shaping

modified Feb 8 at 14:00 nbro♦ 25.4k

3

votes

1

answer

113

views

How can we find the value function by solving a system of linear equations without knowing the policy?

markov-decision-process policies bellman-equations transition-model

modified Feb 16 at 14:47 nbro♦ 25.4k

5

votes

2

answers

53

views

Why is tf.abs non-differentiable in Tensorflow?

tensorflow backpropagation relu gradient

modified Feb 18 at 11:35 nbro♦ 25.4k

3

votes

2

answers

63

views

Is it possible to do face recognition with just the eyes?

machine-learning deep-learning facial-recognition

modified Feb 6 at 10:32 exius 31

3

votes

1

answer

62

views

Why can we take the action $a$ from the next state $s'$ in the max part of the Q-learning update rule, if that action doesn't lead to any reward?

q-learning markov-decision-process off-policy-methods on-policy-methods markov-property

modified Feb 21 at 10:50 nbro♦ 25.4k

1

vote

2

answers

92

views

Does it make sense to combine classifiers trained on the same dataset?

machine-learning classification ensemble-learning

answered 17 hours ago htl 380

2

votes

1

answer

63

views

Would it be possible to determine the dataset a neural network was trained on?

neural-networks machine-learning datasets generative-model

answered Feb 12 at 16:23 nbro♦ 25.4k

1

vote

1

answer

103

views

Is there a convention on the order of multiplication of the weights with the inputs in neural nets?

neural-networks implementation weights

modified Feb 7 at 13:26 nbro♦ 25.4k

1

vote

2

answers

60

views

Is there any situation in which breadth-first search is preferable over A*?

comparison search a-star breadth-first-search

answered Feb 25 at 13:04 Djib2011 2,949

2

votes

1

answer

58

views

Does stochasticity of an environment necessarily mean non-stationarity in MDPs?

reinforcement-learning markov-decision-process

modified Feb 25 at 1:18 nbro♦ 25.4k

3

votes

1

answer

57

views

GAN Generator Output w/ Periodic Noise

neural-networks convolutional-neural-networks generative-adversarial-networks semi-supervised-learning

modified Feb 19 at 15:46 Hazar 61

1

vote

2

answers

93

views

What is the target output for updating a Deep Q Network

neural-networks reinforcement-learning q-learning dqn deep-rl

modified Feb 3 at 10:51 quest ions 82

-2

votes

1

answer

111

views

What is the difference between statistical AI and Classical AI with suitable examples? [duplicate]

machine-learning

modified Feb 4 at 18:03 Basile Starynkevitch 660

2

votes

1

answer

50

views

Can the rewards be matrices when using DQN?

reinforcement-learning dqn reward-functions reward-design multi-objective-rl

modified Feb 3 at 7:30 Neil Slater 17.3k

5

votes

0

answers

78

views

Why did the developement of neural networks stop between 50s and 80s?

neural-networks backpropagation history weights perceptron

modified Feb 26 at 13:58 nbro♦ 25.4k

1

vote

2

answers

60

views

What is the purpose of storing the action $a$ within an experience tuple?

reinforcement-learning experience-replay

modified Feb 22 at 23:53 nbro♦ 25.4k

2

votes

1

answer

58

views

How would I compute the optimal state-action value for a certain state and action?

reinforcement-learning q-learning value-functions bellman-equations

modified Feb 23 at 11:16 nbro♦ 25.4k

3

votes

1

answer

53

views

When using PCA for dimensionality reduction of the feature vectors to speed up learning, how do I know that I'm not letting the model overfit?

machine-learning training overfitting principal-component-analysis dimensionality-reduction

modified Feb 24 at 11:59 nbro♦ 25.4k

1

vote

1

answer

40

views

How to go about classifying 1000 classes?

machine-learning classification text-classification

answered Feb 3 at 15:24 spiridon_the_sun_rotator 301

1

vote

1

answer

40

views

Is it possible to use k-nearest neighbour for classification with more than two attributes?

machine-learning classification k-nearest-neighbors

modified Feb 6 at 14:10 nbro♦ 25.4k

3

votes

1

answer

51

views

How to treat (label and process) edge case inputs in machine learning?

computer-vision classification datasets object-detection yolo

modified Feb 5 at 23:11 juicedatom 457

1

vote

2

answers

51

views

What kind of word embedding is used in the original transformer?

natural-language-processing transformer attention word-embedding

modified Feb 6 at 15:14 nbro♦ 25.4k

3

votes

2

answers

37

views

Accuracy metric for Clustering

unsupervised-learning clustering accuracy

answered Feb 10 at 23:19 Abhishek Verma 367

2

votes

1

answer

61

views

Why do I get the best policy before Q values converge using DQN?

dqn deep-rl convergence value-functions policies

modified Feb 6 at 12:25 nbro♦ 25.4k

1

vote

1

answer

60

views

Is there a graph neural network algorithm that can deal with a different number of input and output nodes?

neural-networks reference-request graph-neural-networks model-request algorithm-request

modified Feb 12 at 22:14 CorruptedHeapScapeGoat 236

1

vote

1

answer

60

views

How exactly is Monte Carlo counterfactual regret minimization with external sampling implemented?

python papers monte-carlo-methods game-theory games-of-chance

answered Feb 12 at 21:26 Phil 31

3

votes

1

answer

60

views

How do you calculate KL divergence on a three-dimensional space for a Variational Autoencoder?

convolutional-neural-networks computer-vision math variational-autoencoder kl-divergence

modified Feb 10 at 12:29 Tinu 535

3

votes

0

answers

74

views

Why does this formula $\sigma^2 + \frac{1}{T}\sum_{t=1}^Tf^{\hat{W_t}}(x)^Tf^{\hat{W_t}}(x_t)-E(y)^TE(y)$ approximate the variance?

bayesian-deep-learning bayesian-neural-networks uncertainty-quantification variance

modified Feb 21 at 0:12 user8714896 385

1

vote

0

answers

71

views

Why does the VAE using a KL-divergence with a non-standard mean does not produce good images?

pytorch variational-autoencoder kl-divergence mnist

modified Feb 19 at 0:16 axon 43

-1

votes

1

answer

54

views

What model to use to get a robust model to predict next 3 days of sales even for products that have just sold once ever?

python time-series regression

modified Feb 7 at 16:04 kjetil b halvorsen 128

0

votes

1

answer

53

views

Extracting information from RNA sequence

neural-networks tensorflow prediction biology

modified Feb 16 at 8:13 yakhyo 134

3

votes

1

answer

53

views

What is Federated Learning?

terminology definitions academia federated-learning

modified Feb 25 at 6:03 Pluviophile 1,041

2

votes

1

answer

41

views

How UCT in MCTS selection phase avoids starvation?

monte-carlo-tree-search upper-confidence-bound

modified Feb 10 at 13:48 nbro♦ 25.4k

2

votes

2

answers

52

views

How do we know that the neurons of an artificial neural network start by learning small features?

neural-networks convolutional-neural-networks deep-neural-networks features feature-learning

modified Feb 26 at 15:14 nbro♦ 25.4k

0

votes

0

answers

66

views

How to make an ensemble model of two LSTM models with different window sizes i.e. different data shapes

python keras long-short-term-memory ensemble-learning

modified Feb 7 at 16:04 Dennis Soemers♦ 8,498

Top Questions

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