Pluto dev confirms the StarCraft agent used zero imitation learning from humans
TimDarcet · x · 2026-10-09
In a discussion over whether the StarCraft agent Pluto used imitation learning, developer TimDarcet confirmed it truly did not learn from human data at all — "no imitation, check the readme." AlphaStar, by contrast, leaned heavily on imitation learning (only the first step), making Pluto's pure self-play RL result notably impressive.
More from Research
- EMNLP paper: diffusion LMs internally encode latent denoising timestep representations — gsarti_ · 2026-10-23
- KU Leuven and VIB open two Group Leader professorships in AI for Biology, start 2027 — jpillowtime · 2026-10-23
- Subspace Tuning curbs LLM catastrophic forgetting by freezing principal singular directions — scottniekum · 2026-10-23
- Perceptic launches PercepticOS, an AI workspace for 'scientific superintelligence' — nathanbenaich · 2026-10-23
- EPFL's Bosselut to talk long context as test-time learning at ELLIS seminar — andre_t_martins · 2026-10-23
- DRATS: Adaptive Task Sampling Boosts Worst-Task Performance in Multi-Task RL — Cohere · 2026-10-23