Differences
This shows you the differences between two versions of the page.
| Next revision | Previous revision | ||
|
courses:rg:2012:sigtest-mt-zilka [2012/11/14 17:19] zilka vytvořeno |
courses:rg:2012:sigtest-mt-zilka [2013/12/02 22:18] (current) popel |
||
|---|---|---|---|
| Line 13: | Line 13: | ||
| Does the bootstrap resampling (Section 5) assume normal (Gaussian) distribution of the scores of samples? | Does the bootstrap resampling (Section 5) assume normal (Gaussian) distribution of the scores of samples? | ||
| ===== Question 4 ===== | ===== Question 4 ===== | ||
| - | We bootstrapped 1000 test sets, computed scoreA-scoreB on each, and we got -1000, | + | We bootstrapped 1000 test sets, computed scoreA-scoreB on each, and we got -1000, |
| Based on Section 6, which system is better - A or B? | Based on Section 6, which system is better - A or B? | ||
| Line 32: | Line 32: | ||
| Can you reformulate Section 4 using this view? What is the observed test statistic and what is the null hypothesis? | Can you reformulate Section 4 using this view? What is the observed test statistic and what is the null hypothesis? | ||
| + | ====== Presentation ====== | ||
| + | * We answered: | ||
| + | * Question 1 - BLEU scores are: 1 - 1.0, 2 - not defined (0.0 or some smoothed value in practice), 3 - 0.2 (based on the incorrect formula in the paper which is missing 1/4) | ||
| + | * Question 2 - broad sampling, samples far apart distributed -> {data_1, data_101, data_201, ...} | ||
| + | |||
| + | ===== Section 3 ===== | ||
| + | * motivation: we don't usually have 30k sentences for testing, so we need an approximate method to obtain reliable scores | ||
| + | * method: divide test set into 100 smaller test sets (300 sentences each) | ||
| + | * consecutive samples - for each of the sets BLEU score varies in range +-8 % | ||
| + | * non-consecutive samples (broad apart) - for each of the sets BLEU varies much less - +-1.5 % | ||
| + | * they make an assumption and claim that there is no difference between comparing output of 2 different MT systems and output of 1 MT systems that is trained just with different data | ||
| + | * Lukas Zilka complained about this assumption - they should have conducted some experiments to support their claim, as there is nothing that suggests we can generalize like that | ||
| + | |||
| + | ===== Section 4, 5 ===== | ||
| + | * we cannot use Student' | ||
| + | * so for estimating the confidence intervals we will use randomized test set generation (= bootstrap resampling) - e.g. we build 1000 new test sets of size 300 sentences out of our small test set of 300 sentences (i.e. we draw (with replacement) samples from the small test set; so we should get 1000 different test sets) | ||
| + | * answer to Question3 - they do not assume there is any particular distribution in the set of BLEU scores of the 1000 test sets (i.e. their method would work regardless of whether the distribution is normal, uniform or any other), but it is perhaps normally distributed | ||
| + | |||
| + | ===== Section 6 ===== | ||
| + | * they use bootstrap resampling to compare 2 systems; we want to determine whether system 1 is better than system 2; we want to determine that from a set of differences of system' | ||
| + | * so we determine in what percent of cases system 1 beats system 2, and that's our final confidence that system 1 is better than system 2 (e.g. 45 times out of 50 -> 90% confidence) | ||
| + | * the rest of the paper just proves that the assumption is correct | ||
| + | |||
| + | ===== Martin' | ||
| + | * two philosophical views of p-value - Fisher' | ||
| + | * we usually set a null hypothesis H0 as: systems are the same, and alternative hypothesis HA: there is difference in the systems; P(H0) + P(HA) = 1 | ||
| + | * p-value = | ||
| + | * P(T(X)> | ||
| + | * unfortunately we tend to view the p-value as P(H0|x) which it is not and we need to apply the Bayes' theorem to get it | ||
| + | * bootstrap resampling can be viewed as p-value = P(d(x) < 0|H0) = P(d(x) > 2*d(x_orig)|H0), | ||