Mutual exclusivity as a challenge for neural networks
share
Strong inductive biases allow children to learn in fast and adaptable ways. Children use the mutual exclusivity (ME) bias to help disambiguate how words map to referents, assuming that if an object has one label then it does not need another. In this paper, we investigate whether or not standard neural architectures have a ME bias, demonstrating that they lack this learning assumption. Moreover, we show that their inductive biases are poorly matched to early-phase learning in several standard tasks: machine translation and object recognition. There is a compelling case for designing neural networks that reason by mutual exclusivity, which remains an open challenge.
READ FULL TEXT