LSTM, ConvLSTM, MDN-RNN and GridLSTM Memory-based Deep Reinforcement Learning.
Fernando Fradique DuarteNuno LauArtur PereiraLuís Paulo ReisPublished in: ICAART (2) (2023)
Keyphrases
- recurrent neural networks
- reinforcement learning
- long short term memory
- function approximation
- feed forward
- neural network
- temporal difference learning
- reinforcement learning algorithms
- model free
- learning algorithm
- state space
- nearest neighbor
- markov decision processes
- information systems
- policy search
- reinforcement learning methods
- temporal difference
- information retrieval
- data sets
- artificial neural networks
- multi agent
- data mining
- optimal policy
- unsupervised learning
- learning process
- control problems
- deep learning
- memory based learning
- multi agent reinforcement learning
- knowledge base
- belief nets
- robotic control
- direct policy search
- search engine