The model wasn't tested in real world environment due to time limitations and student simulators don't model the all of the characteristics of the students. 

San Diego State University

Students often learn huge chunks of information in a limited amount of time, sometimes just a day before an exam, which is ineffective for long-term memorization. Hermann Ebbinghaus was the first to describe the forgetting curve in 1880. **Spaced repetition**, which involves memorizing material at varying intervals, has a positive effect on information retention. One possibility for implementing spaced repetition is through an e-learning environment, where an intelligent agent can adapt and present material to students based on their personal preferences.

Introduction (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)

**Motivation**

1. Popularity of e-learning websites/applications
2. "Recommending content and personalizing the learning sessions is a highly sought after tasks on these platforms"
3. Explore the opportunities to substitute manually selected criteria

**Problem**

4. Deeper explore and extend work Accelerating Human Learning with Deep Reinforcement Learning by several modifications.



Motivation and Problem Description (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)

How does DRL perform compared to other baseline policies when we replace
the reward from being the exact probability of recalling each item,  to a realistically observable reward such as the average of the sum of correct outcomes in a sample of exercises?


What is the effect of replacing the same reward function with an RNN model
that predicts the reward (reward shaping)? We used Long Short Term
Memory (LSTM), a kind of recurrent neural network (RNN), which
has been shown to achieve good results for sequential prediction tasks

Learn Before

Related

Related