online and lightweight kernel-based approximated policy iteration for dynamic p-norm linear adaptive filtering.
Yuki AkiyamaMinh VuKonstantinos SlavakisPublished in: CoRR (2022)
Keyphrases
- lightweight
- adaptive filtering
- policy iteration
- markov decision processes
- filtering method
- reinforcement learning
- speech signal
- optimal policy
- finite state
- model free
- least mean square
- sample path
- linear combination
- temporal difference
- policy evaluation
- fixed point
- optimal control
- neural network
- least squares
- pattern recognition
- support vector
- recursive least squares