-
What are provided as environment parameters
- action space
- knowledge structure
-
What is the action space, knowledge or item?
- knowledge
- item
There are two different mode for
no_measurement_error
with_measurement_error
- Which type is the learner, infinity or finite?
- infinity
- finite
- Which mode is the response of the learner, real or trait?
- real
- trait
-
What are the types of items in this environments?
- learning item
- test item
-
Is the learning item base same with the test item base?
- Yes
- No
-
What are included in an item?
- content
- knowledge
- single
- multiple
- attribute
- guessing
- slipping
-
What is the relation between item and knowledge in learning item?
- one-to-one
- one-to-many
- many-to-one
- many-to-many
-
What is the relation between item and knowledge in test item?
- one-to-one
- one-to-many
- many-to-one
- many-to-many
- Step reward:
$R(t) = \sum_{k=1}^{K}[\alpha_k(t+1)-\alpha_k(t)]$ - Episode reward:
$G=\sum_{t=0}^{T-1}R(t)$
Find these two equations in [1] (Eq.(1) and Eq.(2))
- What is recommended (i.e., what is the action space), knowledge or item?
- knowledge
- item
- binary: Study I in Tang et al. [1]
- tree: Study II in Tang et al. [1]
Tang et al. [1] did not provide source code.
[1] Tang X, Chen Y, Li X, et al. A reinforcement learning approach to personalized learning recommendation systems[J]. British Journal of Mathematical and Statistical Psychology, 2019, 72(1): 108-135.