Kernaut: coding agents design Gaussian process kernels via program search
sirbayes · x · 2026-10-08
A new paper led by @richardcsuwandi introduces Kernel Autoresearch (Kernaut), which treats kernel design as open-ended program synthesis: coding agents write kernels as programs, going beyond prior work limited to combining base kernels via sums, products, and neural embeddings from a fixed library. Construction contracts ensure accepted kernels remain valid (e.g., PSD).
Related event: Kernaut: Coding Agents Autodiscover Interpretable GP Kernels(3 posts)→
More from coding & agent
- LangChain revamps Deep Agents skills as teams adopt them as the standard for agent domain knowledge — LangChain · 2026-10-08
- 280+ generations to make Superman vs Saitama: MiniMax H3 workflow lessons — Automatic-Building84 · 2026-10-08
- mitchellh Calls His OpenAI Dot to Move Benchmarks Across Machines via Tailscale — charliermarsh · 2026-10-08
- A Developer's Map of the Data Tooling Landscape: Ingestion, Storage, Modeling, Governance — bibryam · 2026-10-08
- Operator-1 launches as autonomous document extraction agent for long tables and sprawling PDFs — ycombinator · 2026-10-08
- good-css: 47 modern CSS techniques that make coding agents actually use :has() and subgrid — mattpocockuk · 2026-10-08