Space and Time in Mind
The generation of cognitive maps can be performed using VAEGAN (variational autoencoder + generative adversarial network) or a deep neural network with a recurrent loop. This research shows how the cognitive map is self-organized. This project is to understand what it means to spatialize time in the cognitive system. This is a DNN reconstruction of the previous studies of modeling "the others" using a recurrent neural network.
Hiroki Kojima and Takashi Ikegami. "Organization of a Latent Space structure in VAE/GAN trained by navigation data." arXiv preprint arXiv:2102.01852 (2021).
Takashi Ikegami, Gentaro Morimoto Chaotic Itinerancy in Coupled Dynamical Recognizers. CHAOS 13:1133-1147, 2003