Question Answering On Babi

Mean Error Rate

평가 결과

이 벤치마크에서 각 모델의 성능 결과

비교 표

모델 이름	Mean Error Rate
finding-remo-related-memory-object-a-simple	1.2%
dynamic-neural-turing-machine-with-soft-and	-
tracking-the-world-state-with-recurrent	9.7%
recurrent-relational-networks	0.46%
scaling-memory-augmented-neural-networks-with	6.4%
self-assttentive-associative-memory	0.39%
neural-stored-program-memory	5.6%
rotational-unit-of-memory	-
gated-orthogonal-recurrent-units-on-learning	-
end-to-end-memory-networks	7.5%
h-mem-harnessing-synaptic-plasticity-with	-
memory-enriched-computation-and-learning-in	-
scaling-memory-augmented-neural-networks-with	28.7%
query-reduction-networks-for-question	0.3%