论文信息 - Holographic Recurrent Networks

Holographic Recurrent Networks

Holographic Recurrent Networks (HRNs) are recurrent networks which incorporate associative memory techniques for storing sequential structure. HRNs can be easily and quickly trained using gradient descent techniques to generate sequences of discrete outputs and trajectories through continuous space. The performance of HRNs is found to be superior to that of ordinary recurrent networks on these sequence generation tasks.

Tony Plate | T. Plate

[1] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[2] Terrence J. Sejnowski,et al. NETtalk: a parallel network that learns to read aloud , 1988 .

[3] Yann LeCun,et al. Reverse TDNN: An Architecture For Trajectory Generation , 1991, NIPS.

[4] Tony Plate,et al. Holographic Reduced Representations: Convolution Algebra for Compositional Distributed Representations , 1991, IJCAI.

[5] J. Elman. Distributed Representations, Simple Recurrent Networks, And Grammatical Structure , 1991 .

[6] Jeffrey L. Elman,et al. Distributed Representations, Simple Recurrent Networks, and Grammatical Structure , 1991, Mach. Learn..

[7] Andrew S. Noetzel,et al. Forcing Simple Recurrent Neural Networks to Encode Context , 1992 .