A pulse neural network reinforcement learning algorithm for partially observable Markov decision processes

Mathematics

• Partially Observable Markov Decision Process 100%
• Reinforcement Learning 90%
• Learning Algorithm 71%
• Neuron 62%
• Neural Networks 57%
• Time series 23%
• Reinforcement 22%
• Decay Rate 17%
• Learning 15%
• Internal 14%
• Feedforward Networks 13%
• Computer Experiments 10%
• Evaluation Function 10%
• Detector 10%
• Coincidence 9%
• Discrimination 9%
• Decay 7%
• Engineering 7%
• Verify 7%
• Output 6%
• Model 5%
• Modeling 5%

Engineering & Materials Science

• Markov processes 64%
• Reinforcement learning 60%
• Neurons 55%
• Learning algorithms 50%
• Neural networks 36%
• Time series 22%
• Reinforcement 13%
• Function evaluation 9%
• Detectors 6%