Posts by Tag

• policy gradient 2
• expected reward 2
• sampling 2
• MRT 2
• REINFORCE 2
• neural machine translation 2
• semantic parsing 2
• exposure bias 2
• reinforcement learning 2
• score function gradient estimator 1
• monolingual data 1
• future challenges 1
• BLEU 1

policy gradient

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

expected reward

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

sampling

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

MRT

Counterfactual Learning of Semantic Parsers When Even Gold Answers Are Unattainable

less than 1 minute read

How can we train semantic parsers if neither question-parse nor question-answer pairs can be collected?

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

REINFORCE

Counterfactual Learning of Semantic Parsers When Even Gold Answers Are Unattainable

less than 1 minute read

How can we train semantic parsers if neither question-parse nor question-answer pairs can be collected?

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

neural machine translation

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

semantic parsing

Counterfactual Learning of Semantic Parsers When Even Gold Answers Are Unattainable

less than 1 minute read

How can we train semantic parsers if neither question-parse nor question-answer pairs can be collected?

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

exposure bias

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

reinforcement learning

Counterfactual Learning of Semantic Parsers When Even Gold Answers Are Unattainable

less than 1 minute read

How can we train semantic parsers if neither question-parse nor question-answer pairs can be collected?

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Back to Top ↑

score function gradient estimator

Taming Wild Reward Functions: The Score Function Gradient Estimator Trick

less than 1 minute read

This post explains the need for the score function gradient estimator trick and how it works.

Back to Top ↑

monolingual data

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Back to Top ↑

future challenges

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Back to Top ↑

BLEU

RL in NMT: The Good, the Bad and the Ugly

less than 1 minute read

Discussing good, bad and ugly practices of reinforcement learning in neural machine translation.

Back to Top ↑