Learning to Control Rapidly Changing Synaptic Connections: An Alternative Type of Memory in Sequence Processing Artificial Neural Networks.
Kazuki IrieJürgen SchmidhuberPublished in: CoRR (2022)
Keyphrases
- rapidly changing
- artificial neural networks
- learning rules
- learning systems
- neural network
- learning process
- robot control
- supervised learning
- knowledge acquisition
- learning algorithm
- using artificial neural networks
- autonomous robots
- learning tasks
- data processing
- active learning
- decision trees
- memory requirements
- feed forward
- computational intelligence
- computational power
- control system
- adaptive control
- random access
- past experience
- reinforcement learning
- motor control
- control rules