Xavier Hinaut – Towards interactive models with Reservoir Computing
Dr. Xavier Hinaut
INRIA National Institute for Research in Digital Science and Technology
Before the lecture, please read the first two publications listed in the Literature section (the third one is optional). If you want to better understand the subject of reservoir computing, there are additional materials, suggested by Dr. Hinaut.
Literature:
Cross-situational learning: Juven, A., & Hinaut, X. (2020). Cross-situational learning with reservoir computing for language acquisition modelling.
Birdsong model: Pagliarini, S., Leblois, A., & Hinaut, X. (2021). Canary Vocal Sensorimotor Model with RNN Decoder and Low-dimensional GAN Generator
Additionally: Hinaut, X. & Dominey, PF. (2013). Real-time parallel processing of grammatical structure in the fronto-striatal system: A recurrent network simulation study using reservoir computing.
A quick overview:
-
A tutorial about implementation: Lukoševičius, M. (2012). A practical guide to applying echo state networks.
An online algorithm with chaotic reservoirs (that also work with non chaotic reservoir) which has a nice description for pattern generation: Sussillo, D., & Abbott, LF. (2009). Generating coherent patterns of activity from chaotic neural networks.
We had a nice discussion concerning reservoir computing, cross-situational learning and the possibility to model interactions and descriptive/imperative language use in that way. Some random thoughts about reservoir computing.