Reservoir computing bootcamp-From Python/NumPy tutorial for the complete beginners to cutting-edge research topics of...

Area of Science:

• Machine Learning
• Computational Neuroscience
• Dynamical Systems Theory

Background:

• Reservoir computing (RC) leverages recurrent neural networks and intrinsic dynamics for time-series processing.
• Physical reservoir computing (PRC) utilizes physical systems as reservoirs, expanding RC applications.
• Interdisciplinary adoption of RC and PRC can foster novel research avenues.

Purpose of the Study:

• Introduce "RC bootcamp," a Jupyter Notebook-based educational resource for learning RC and PRC.
• Facilitate efficient training for collaborators and students, enabling independent experimentation.
• Provide accessible learning materials for individuals with diverse academic backgrounds.

Main Methods:

• Utilizes Python/NumPy for foundational computer science and numerical computation.
• Covers fundamental RC implementations like echo state networks and linear regression.
• Explores dynamical systems theory indicators (Lyapunov exponents, echo state property index, information-processing capacity) and advanced chaotic methods (FORCE learning, innate training, attractor design).

Main Results:

• RC bootcamp is publicly available under an open-source license.
• The material is designed for self-directed learning and practical application.
• Covers a comprehensive range of topics from basic concepts to cutting-edge research.

Conclusions:

• RC bootcamp is a valuable educational tool for mastering RC and PRC.
• The resource is expected to stimulate further research and innovation in the field.
• Empowers learners to conduct their own experiments and contribute to RC/PRC advancements.