Posts about speculation
- Why your active learning algorithm may not do better than random
- Generic recommendations for cheminformatics models.
- Conceptual confusion about desirable outputs of reaction prediction models.
- Why don't ML conferences provide reviewer instructions?
- Is offline model-based optimization a realistic problem? (I'm not convinced)
- Reaction model scores are CRITICAL to multi-step retrosynthesis.
- What ML researchers and users get wrong: optimistic assumptions
- Review of NeurIPS 2024 and predictions for ML in 2025
- Rules of scientific English writing for an international audience.
- When should you expect Bayesian optimization to work well?
- What can eduroam teach us about building research infrastructure
- Scientific conferences as approximate Bayesian inference
- Advice for applying for a PhD in AI for science in 2024
- Being 'data-driven' does not mean that you should use bad data.
- How to guess a Kanji's on-yomi in 4 easy steps