Resources

エリート的な文章の書き方 (Elite Writing Style)

A Japanese-language guide that defines "elite writing" as prose whose paragraphs convey arguments clearly and whose sentences never burden the reader. A practical manual for columnists, essayists, and researchers who want to be genuinely understood.

The importance of stupidity in scientific research

A short essay by Martin A. Schwartz (Journal of Cell Science, 2008) arguing that feeling confused or incompetent in research is not failure but evidence of tackling genuinely unknown questions. Reminds us that "our ignorance is infinite," and that comfort with uncertainty is the foundation for meaningful discovery. Highly recommended for anyone pursuing research.

Supervised Machine Learning for Science

An open online textbook (by Christoph Molnar & Timo Freiesleben) exploring how to apply supervised machine learning in scientific research. Covers the theory and practice of integrating ML not merely as a prediction tool but as a scientific instrument equipped with interpretability, causal reasoning, and uncertainty quantification. Published under CC BY-NC-SA 4.0.

Interpretable Machine Learning

A comprehensive guide to making black-box models explainable (by Christoph Molnar). Covers a wide range from interpretable models such as decision trees and linear regression to model-agnostic methods including LIME and SHAP (Shapley values). The third edition (2024) adds new techniques. Accessible without extensive mathematical prerequisites.

The OFE Comic — Transforming Agriculture

A 28-slide illustrated narrative introducing On-Farm Experimentation (OFE), an approach that bridges researchers' scientific knowledge with farmers' field expertise — grounded in the principle that "every farm is its own equation." Published by GOFEN (Global On-Farm Experimentation Network), led by Cornell University in collaboration with Université de Montpellier, the Farmers Datalab, and the European Commission. Freely available in nine languages.

Neural Network Console

A Japanese-language YouTube channel by Sony that teaches deep learning from the ground up. It explains how neural networks work and the ideas behind training them through clear visuals rather than heavy math. A great starting point for students new to deep learning or anyone who wants an intuitive grasp of the concepts.