Cold-Start Reinforcement Learning with Softmax Policy Gradients.
Nan DingRadu SoricutPublished in: CoRR (2017)
Keyphrases
- cold start
- reinforcement learning
- optimal policy
- action selection
- recommender systems
- temporal difference learning
- markov decision process
- cold start problem
- collaborative filtering
- data sparsity
- function approximators
- state space
- reward function
- function approximation
- user preferences
- tag recommendation
- personalized recommendation
- markov decision processes
- transfer learning
- implicit feedback
- data sparseness
- temporal difference
- reinforcement learning algorithms
- user interface