Double minMerge
Minimum similarity required between two senses if they are to be merged.

Double modelPriorWeight
Magnitude of preference for merging relative to preserving likelihood.
(Note: this is like the alpha parameter from the Part 4 handout.)

Integer batchSize
Number of new instances of a word to accumulate between merging episodes.
(Note: the Part 4 handout uses on "online" version of the algorithm,
in which data instances are incorporated one at a time.)

Integer trainingPasses
Replicate training data this many times.

Double minLabel
Minimum P(word | action) required to emit word.

Double minObey
Minimum P(sense | worldstate) require to use sense in setting linking Fstruct.
(Note: linking Fstruct refers to either Motor or World feature structure.)

(Note: These features are related to more advanced features of the
VerbLearn program. You may experiment with these if you like,
but you should stick to the basic training/testing parameters
above for the purposes of Part 2 of the assignment.)

Double minExplain
Minimum P(feature | sense) required to explain away feature.
(Note: this parameter is used only in multislot learning.)

Double minSetFeature
Minimum "peakedness" of distribution required to set feature in linking Fstruct.
(Note: linking Fstruct refers to either Motor or World feature structure.

(Note: the following three parameters involve "virtual samples", which
allow varying amounts of additional data to be incorporated
based on seen data.)

Boolean adaptVirtuals
Whether to adapt Dirichlet prior in each slot during learning.

Double maxVirtual
Prevent uniform distribution from causing infinite virtuals.

Double virtualInertia
Effective number of samples supporting current virtuals when adapting them.