Numerai Grandmaster: a simple, repeatable pipeline beat complex architectures
matlabulous · x · 2026-10-23
Episode 6 of Numerai Council of Elders' "Out of Sample" series features 2025 Numerai Grandmaster Bruno Guberfain, who found after years of experimenting with elaborate systems that a streamlined, highly repeatable pipeline — and keeping model freshness — was what actually drove his performance, more than complex architectures.
More from Research
- arXiv: Self-Evolving Agents through Privileged Retrospective Distillation (Meta AI) — rohanpaul_ai · 2026-10-23
- Meta's RSD paper: agents self-improve from unlabeled runs, lifting Qwen3-14B success from 12.9% to 20.6% — rohanpaul_ai · 2026-10-23
- Meta Expert: cross-layer expert sharing matches vanilla MoE with 70-80% experts — pmttyji · 2026-10-23
- 10 of 12 Beta Robots Worked Without Fine-Tuning — Do We Still Need It? — micoolcho · 2026-10-23
- EmbodiedEvo lets robot agents self-train, banking reusable Skills and Lessons across tasks — jiqizhixin · 2026-10-23
- PACE: chunked 4-bit recurrent-state quantization matches 8-bit at 2/3 memory — pmttyji · 2026-10-23