A Bio-Inspired Recurrent Neural Network with Self-Adaptive Neurons and PCM Synapses for Solving Reinforcement Learning Tasks.
Stefano BianchiI. Muñoz-MartínS. HashemkhaniGiacomo PedrettiDaniele IelminiPublished in: ISCAS (2020)
Keyphrases
- learning tasks
- bio inspired
- neural models
- recurrent neural networks
- hebbian learning
- artificial neural networks
- neural model
- hidden layer
- neural network
- swarm intelligence
- machine learning
- learning algorithm
- feed forward
- transfer learning
- machine learning algorithms
- spiking neural networks
- reinforcement learning
- recurrent networks
- supervised learning
- multi task
- learning experience
- receptive fields
- back propagation
- kernel methods
- visual cortex
- multi label
- genetic algorithm
- neural network structure
- search space
- multilayer perceptron
- optimization problems
- training data
- activation function
- image segmentation
- input patterns
- evolutionary computation