Blog6 posts
Blog
Paper TLDRs in deep learning, mathematics, et cetera.
Series
- P1Global-Scale Forest Height Estimation
- P2Capturing Temporal Dynamics in Canopy Height
- P3ECHOSAT: Canopy Height Over Space and Timesoon
All posts
- Preprint June 15, 2026 7 min read
What Do Evolutionary Coding Agents Evolve?
A closer look at what changes inside evolutionary code search runs.
- ICML26 Workshop Oral April 12, 2026 9 min readAI for Mathematical Discovery · part 2
The Agentic Researcher: A Practical Guide to AI-Assisted Research in Mathematics and Machine Learning
An open-source framework that turns CLI coding agents into autonomous research assistants.
- ICML25 May 30, 2025 13 min read
Approximating Latent Manifolds in Neural Networks via Vanishing Ideals
Describing class manifolds in a network's latent space with polynomials, and replacing layers with them.
- ICML25 May 29, 2025 13 min readAI for Mathematical Discovery · part 1
Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?
Neural networks as a tool for finding colorings of the plane, including two new 6-colorings.
- ICML25 May 1, 2025 8 min readTree Canopy Height Estimation · part 2
Capturing Temporal Dynamics in Tree Canopy Height
From a single-year map to canopy height over time: a 10 m map of Europe for 2019 to 2022.
- ICML24 April 13, 2025 10 min readTree Canopy Height Estimation · part 1
Global-Scale Forest Height Estimation
A global canopy height map for 2020, learned from satellite imagery and sparse GEDI lidar measurements.