Embodiment dictates learnability in neural controllers.
Joshua P. PowersRyan GrindleSam KriegmanLapo FratiNick CheneyJosh C. BongardPublished in: CoRR (2019)
Keyphrases
- evolutionary robotics
- network architecture
- neural network
- inductive inference
- neural model
- dnf formulas
- control system
- finite automata
- boolean functions
- learning algorithm
- reinforcement learning
- learning rules
- biologically plausible
- hebbian learning
- neural fuzzy
- vapnik chervonenkis dimension
- uniform distribution
- bio inspired
- learning theory
- control strategy
- controller design
- cognitive science
- positive and negative
- uniform convergence
- agnostic learning
- sufficient conditions