Hottest Questions Today

0

votes

0

answers

14

views

Reinforcement Learning: How to train policy model continuously the incremental way

modified 4 hours ago datdinhquoc 753

0

votes

0

answers

21

views

How to think and build an AI project?

neural-networks python optimization

modified 10 hours ago Djib2011 2,934

1

vote

0

answers

13

views

Can someone explain to me this implementation of Tile Coding using Hash Tables?

reinforcement-learning python implementation function-approximation tile-coding

modified 5 hours ago nbro♦ 23.7k

0

votes

1

answer

16

views

How to make input variable as trainable parameter in a neural network?

neural-networks optimization inverse-rl

modified 10 hours ago nbro♦ 23.7k

0

votes

1

answer

38

views

What is the difference between a fitness function and a reward function?

reinforcement-learning loss-functions evolutionary-algorithms reward-functions fitness-functions

modified 13 hours ago nbro♦ 23.7k

1

vote

0

answers

13

views

Understanding policy gradient derivation with importance sampling

reinforcement-learning policy-gradients off-policy-methods

asked 15 hours ago Flight Booker 11

3

votes

0

answers

39

views

How to implement REINFORCE with eligibility traces?

reinforcement-learning actor-critic-methods reinforce eligibility-traces

modified 20 hours ago nbro♦ 23.7k

1

vote

1

answer

22

views

Bounding Box Regression - An Adventure in Failure

tensorflow object-detection weights-initialization bounding-box

modified 6 hours ago nbro♦ 23.7k

3

votes

0

answers

38

views

When do two identical neural networks have uncorrelated errors?

neural-networks proofs statistics ensemble-learning correlation

modified 15 hours ago EmmanuelMess 169

2

votes

0

answers

23

views

Are there any new weight initialization techniques for DNN published after 2015?

deep-learning reference-request deep-neural-networks weights weights-initialization

modified 11 hours ago nbro♦ 23.7k

1

vote

0

answers

28

views

Understanding Batch Normalization for CNNs

convolutional-neural-networks batch-normalization

modified yesterday IntegrateThis 141

2

votes

1

answer

51

views

How does replacing states with latent representations help RL agents?

reinforcement-learning papers autoencoders generalization state-spaces

modified 21 hours ago nbro♦ 23.7k

1

vote

0

answers

32

views

In UCB, is the actual upper bound an upper bound of an one-sided or two-sided confidence interval?

machine-learning reinforcement-learning multi-armed-bandit upper-confidence-bound

modified yesterday nbro♦ 23.7k

0

votes

0

answers

25

views

+50

Predicting continous value with CNN (prediction of fruit maturity)

convolutional-neural-networks image-recognition continuous-tasks

modified 14 hours ago Antoine Labelle 101

2

votes

1

answer

86

views

How to prove the formula of eligibility traces operator in reinforcement learning?

reinforcement-learning proofs bellman-equations eligibility-traces

modified yesterday nbro♦ 23.7k

0

votes

1

answer

23

views

What are acting as weights in a convolution neural network?

convolutional-neural-networks weights

modified yesterday nbro♦ 23.7k

1

vote

0

answers

18

views

What is the state-space complexity of Spades?

game-ai games-of-chance space-complexity

modified 2 days ago Cohensius 141

1

vote

1

answer

27

views

How to compute the gradient of the cross-entropy loss function with respect to the parameters with softmax activation function?

deep-learning backpropagation math cross-entropy softmax

answered 3 hours ago KID 1

0

votes

1

answer

43

views

When calculating the cost in deep Q-learning, do we use both the input and target states?

reinforcement-learning q-learning dqn loss-functions temporal-difference-methods

modified 2 days ago David Ireland 2,818

-1

votes

0

answers

22

views

The policy gradient method is constantly updated to positive values [closed]

reinforcement-learning policy-gradients

asked 2 days ago ground clown 93

1

vote

0

answers

16

views

In this implementation of pix2pix, why are the weights for the discriminator and generator losses set to 1 and 100 respectively?

loss-functions generative-adversarial-networks pix2pix

modified yesterday nbro♦ 23.7k

1

vote

0

answers

12

views

How are the parameters $\alpha_i$ of hard attention trained?

computer-vision training papers attention

modified yesterday nbro♦ 23.7k

2

votes

0

answers

22

views

Is better to reward short- or long-term progress in Q-learning?

reinforcement-learning q-learning dqn reward-design reward-functions

modified 2 days ago nbro♦ 23.7k

0

votes

3

answers

70

views

How to embed/deploy an arbitrary machine learning model on microcontrollers?

machine-learning tensorflow models

modified 2 days ago Hattori 41

2

votes

1

answer

59

views

Why is neural networks being a deterministic mapping not always considered a good thing?

neural-networks bayesian-deep-learning bayesian-neural-networks

modified Jan 17 at 11:51 nbro♦ 23.7k

1

vote

0

answers

41

views

+50

How do I learn the value function for a POMDP with a single-step horizon (bandit)?

reinforcement-learning value-functions multi-armed-bandit pomdp contextual-bandit

modified 20 hours ago nbro♦ 23.7k

0

votes

0

answers

42

views

How to get started with AI research on detecting human poses? [closed]

research getting-started pose-estimation

modified Jan 17 at 19:31 user79161 341

2

votes

1

answer

30

views

Current research on Gödel machines

reference-request agi problem-solving godel-machine

modified Jan 17 at 16:11 nbro♦ 23.7k

1

vote

0

answers

64

views

Is object-based representation of the observation space feasible?

reinforcement-learning python environment state-spaces

modified 2 days ago Shahin 153

0

votes

0

answers

14

views

Conversion of convolution filter weights between NCHW and NHWC formats

tensorflow pytorch convolution

modified Jan 18 at 6:17 Shayan Shafiq 127

0

votes

1

answer

49

views

+100

How to predict the best from a set of messages - best practice

machine-learning training data-preprocessing feature-selection features

modified 2 days ago Daniel B. 517

0

votes

0

answers

16

views

Keras 1D CNN always predicts the same result even if accuracy is high on training set

deep-learning convolutional-neural-networks python keras 1d-convolution

modified Jan 17 at 18:46 raulfilipe127 1

1

vote

1

answer

33

views

What happens if there is no activation function in some layers of a neural network?

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

answered Jan 16 at 22:12 David Hoelzer 175

0

votes

1

answer

57

views

Can AI be understood as a generalized statistics tool? [duplicate]

machine-learning comparison definitions statistical-ai statistics

modified Jan 16 at 17:02 nbro♦ 23.7k

0

votes

1

answer

22

views

Why won't my model train with CTC loss?

neural-networks deep-learning tensorflow long-short-term-memory ctc-loss

answered Jan 17 at 15:40 Cameron Martin 101

0

votes

0

answers

14

views

Text generation with LSTM with multiple correlated inputs

natural-language-processing long-short-term-memory

asked Jan 17 at 12:36 AugustinLouisCauchy 1

1

vote

1

answer

30

views

What is the smoothness assumption in SVMs?

machine-learning terminology papers support-vector-machine

answered yesterday Chris Young 36

3

votes

0

answers

90

views

+50

Is it really possible to create the “Perfect Cylinder” used in Universal Approximation Theorem for 1-hidden layer Neural Network?

neural-networks feedforward-neural-network universal-approximation-theorems

modified 2 days ago KoKlA 81

0

votes

0

answers

15

views

Architecture for Logistic Regression with Arbitrary Number of Options

neural-networks machine-learning logistic-regression

asked Jan 16 at 23:14 bluekaterpillar 41

1

vote

0

answers

26

views

Difference in average rewards between taking random actions and following random policies

reinforcement-learning gym

asked Jan 16 at 8:02 artils1997 11

2

votes

0

answers

22

views

How to train/update neural networks faster without a decrease in performance?

neural-networks machine-learning training efficiency

modified Jan 16 at 10:13 Lerner Zhang 499

1

vote

0

answers

19

views

Python catch the reference between 2 strings [closed]

natural-language-processing python chat-bots natural-language-understanding intelligent-personal-assistants

modified Jan 16 at 12:59 nbro♦ 23.7k

-1

votes

0

answers

17

views

Learning Agents I can talk to [closed]

intelligent-agent

asked Jan 16 at 8:49 Michael Price 1

3

votes

2

answers

41

views

Is pre-processing used in deep learning?

machine-learning deep-learning data-preprocessing

answered Jan 14 at 19:46 Alex 111

1

vote

0

answers

21

views

Offline/Batch Reinforcement Learning: when to stop training and what agent to select

training q-learning off-policy-methods batch-learning

modified 2 days ago MetaHG 11

0

votes

0

answers

15

views

How to get the prediction probability of random sample image from multiclass classification model?

convolutional-neural-networks probability multi-label-classification

asked Jan 16 at 5:54 anil 1

1

vote

0

answers

13

views

feeding a NN with tensors with varying spatial dimensions

pytorch image-segmentation fully-convolutional-networks

asked Jan 16 at 9:46 David S 148

3

votes

1

answer

43

views

Can a convolutional network predict states for a RL Agent

reinforcement-learning convolutional-neural-networks

answered Jan 14 at 11:33 David Ireland 2,818

Top Questions

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