Imparting universal linguistic inductive biases via meta-learning

Language description:
Constraint ranking:
Onset, NoCoda, Max » Dep Onset, NoCoda, Dep » Max Onset, Dep » Max » NoCoda NoCoda, Dep » Max » Onset Onset, Max » Dep » NoCoda NoCoda, Max » Dep » Onset Dep » Max » Onset, NoCoda Max » Dep » Onset, NoCoda
Set of consonants:
Set of vowels:
Consonant for insertion:
Vowel for insertion:
Task to be learned:
Training examples seen: 0

Test set predictions
Input
Correct output
Meta-initialized
model's output
Randomly-initialized
model's output
rOau
.rO.a.u.
.rO.a.u.
.rO.a.u.
axxaO
.a.xa.O.
.a.xa.O.
.a.xa.O.
rxxa
.xa.
.xa.
.xa.
axrxu
.a.xu.
.a.xu.
.a.xu.
ttxaO
.xa.O.
.xa.O.
.xa.O.
rOau
.rO.a.u.
.rO.a.u.
.rO.a.u.





Generalization

Use multiple tabs: Click on the type of generalization you want to study.

Sandbox

People can enter their own training set and test set and see how the model learns it.