Results Comparing Likelihood-Based and Average-of-Sum-of-Outcomes Reward Functions (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)
The study compared likelihood-based and average-of-sum-of-outcomes reward functions across the EFC, HLR, and GPL student models. The two reward functions produced relatively similar performance with EFC and HLR, whereas performance fluctuated with GPL and their reward distributions differed. For the GPL student model, the average-of-sum-of-outcomes function performed slightly better than the likelihood-based function (). The average-of-sum-of-outcomes formulation was intended to reduce dependence on modeled student states.
0
1
Contributors are:
Who are from:
Tags
Data Science
Related
Performance of DRL agent when the number of items are varied (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)
Performance of TRPO with reward shaping (research objective) (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)
Observed Relation Between Rewards and Thresholds (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)
Comparison of Performance of TRPO and TNPG algorithms (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)
Results Comparing Likelihood-Based and Average-of-Sum-of-Outcomes Reward Functions (Using deep reinforcement learning for personalizing review sessions on e-learning platforms with spaced repetition)