Posts about machine learning
- Behaviour-based evaluations in Bayesian optimization
- Rebranding BO away from "black-box" and towards "model-based"
- Clarifying noise vs model misspecification in Gaussian Process models (and its importance in BO)
- We are underselling the modularity of Bayesian optimization
- Why I don't care about toy benchmarks in BO
- We have forgotten about utility functions in BO (whoops!)
- My model-centric view of Bayesian optimization
- Predictions for ML/AI in 2026 (and 2025 predictions re-visited).
- Review of NeurIPS 2025
- Finally, an ML conference review guide!
- Justifying expected utility maximization from first principles (Von Neumann–Morgenstern).
- Latent Space COWBOYS: a VAE-BO method I can actually buy into!
- An accessible introduction to mutual information using a d20
- My review guide for machine learning conference papers
- Problems with ML for toxicity prediction.
- Chebyshev Scalarization Explained
- Taking the V out of VAEs: long live KL-regularized autoencoders!
- Coding python packages with AI
- 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.
- Punishing poor reviewers at CVPR
- Why don't ML conferences provide reviewer instructions?
- Is offline model-based optimization a realistic problem? (I'm not convinced)
- Stock responses about statistical significance for reviewing machine learning papers
- Reaction model scores are CRITICAL to multi-step retrosynthesis
- Double checking that Gauche's fingerprint kernels are positive definite
- Review of NeurIPS 2024 and predictions for ML in 2025
- When should you expect Bayesian optimization to work well?
- Thoughts on Google Vizier
- Problems with the top-k diversity metric for diverse optimization
- Sparse Matrices: 5 Tips and Tricks
- An Overview of Gradient Boosting and Popular Libraries for it.
- Turning Adam Optimization into SGD