Future Trajectory Prediction via RNN and Maximum Margin Inverse Reinforcement Learning.
Dooseop ChoiTaeg-Hyun AnKyounghwan AhnJeongdan ChoiPublished in: ICMLA (2018)
Keyphrases
- maximum margin
- inverse reinforcement learning
- hyperplane
- support vector
- structured prediction
- markov networks
- learning algorithm
- nearest neighbor
- support vector machine
- pattern classification
- maximum likelihood
- high order
- multi task
- preference elicitation
- markov random field
- learning tasks
- multiple kernel learning
- hidden markov models
- temporal difference
- reward function
- training set