Deep reinforcement learning (DRL) which involved reinforcement learning and artificial neural network allows agents to take the best possible actions to achieve goals. Spiking Neural Network (SNN) faced difficulty in training due to the non-differentiable spike function of spike neuron. In order to overcome the difficulty, Deep Q network (DQN) and Deep Q learning with normalized advantage function (NAF) are proposed to interact with a custom environment. DQN is applied for discrete action space whereas NAF is implemented for continuous action space. The model is trained and tested to validate its performance in order to balance the firing rate of excitatory and inhibitory population of spike neuron by using both algorithms. Training results showed both agents able to explore in the custom environment with OpenAI Gym framework. The trained model for both algorithms capable to balance the firing rate of excitatory and inhibitory of the spike neuron. NAF achieved 0.80% of the average percentage error of rate of difference between target and actual neuron rate whereas DQN obtained 0.96%. NAF attained the goal faster than DQN with only 3 steps taken for actual output neuron rate to meet with or close to target neuron firing rate.

Deep reinforcement learning which involved reinforcement learning with artificial neural networks allows an agent to take the best possible actions in a virtual environment to achieve goals. Spike neuron has a non-differentiable spike generation function that caused SNN training faced difficulty. In order to overcome the difficulty, Deep Q Network (DQN) is proposed to act as an agent to interact with a custom environment. A spike neuron is modelled by using NEST simulator. Rewards are given to the agent for every action taken. The model is trained and tested to validate the performance of the trained model in order to attain balance the firing rate of excitatory and inhibitory population of spike neuron. Training result showed the agent able to handle the environment. The trained model capable to balance the excitation and inhibition of the spike neuron as the actual output neuron rate is close to or same with the target neuron firing rate. The average percentage error of rate of difference between output and target neuron rate for 5 episodes achieved 0.80%.