Differences
This shows you the differences between two versions of the page.
Next revision | Previous revision | ||
courses:rg:2013:convolution-kernels [2013/02/26 10:03] dusek vytvořeno |
courses:rg:2013:convolution-kernels [2013/03/12 11:27] (current) popel <latex>x</latex> was not rendered |
||
---|---|---|---|
Line 11: | Line 11: | ||
- Find an error in one of the formulae in the paper. | - Find an error in one of the formulae in the paper. | ||
+ | ==== Answers ==== | ||
+ | - | ||
+ | * **Generative models** use a two-step setup. They learn class-conditional (likelihood) < | ||
+ | * they learn the joint distributions: | ||
+ | * They learn more than is actually needed, but are not prone to partially missing input data. | ||
+ | * They are able to " | ||
+ | * Examples: Naive Bayes, Mixtures of Gaussians, HMM, Bayesian Networks, Markov Random Fields | ||
+ | * **Discriminative models** do everything in one-step -- they learn the posterior < | ||
+ | * They are simpler and can use many more features, but are prone to missing inputs. | ||
+ | * Examples: SVM, Logistic Regression, Neural network, k-NN, Conditional Random Fields | ||
+ | - Each CFG rule generates just one level of the derivation tree. Therefore, using " | ||
+ | * '' | ||
+ | * It could be modelled with an augmentation of the nonterminal labels. | ||
+ | * CFGs can't generate non-projective sentences. | ||
+ | * But they can be modelled using traces. | ||
+ | - The derivation is actually quite simple: | ||
+ | - < | ||
+ | - < | ||
+ | - < | ||
+ | - < | ||
+ | - < | ||
+ | - Convolution is defined like this: < | ||
+ | - There is a (tiny) error in the last formula of Section 3. You cannot actually multiply tree parses, so it should read: < | ||
+ | |||
+ | ==== Report ==== | ||
+ | |||
+ | We discussed the answers to the questions most of the time. Other issues raised in the discussion were: | ||
+ | |||
+ | * **Usability** -- the approach is only usable for // | ||
+ | * **Scalability** -- they only use 800 sentences and 20 candidates per sentence for training. We believe that for large data (milions of examples) this will become too complex. | ||
+ | * **Evaluation** -- it looks as if they used a non-standard evaluation metric to get " |