PD-NCA AutoLab: Self-Organizing Neural Intelligence as a Testbed for Autonomous Science

David Scott Lewis, Enrique Zueco

Submission Type: I want my submission to be considered for both oral and poster presentation.

Keywords: autonomous science, agents and autonomous AI researchers, self-driving labs, neural cellular automata, self-organizing systems, virtual laboratories, benchmark environments, world models, causal discovery, intervention planning, reversal planning, open-ended discovery, differentiable simulation, artificial life, mechanism recovery, resilience modeling, reproducibility, auditability, unconventional computing, AI for biology, safe AI, datasets benchmarks and software, governance, ML algorithmic advances

TL;DR: PD-NCA AutoLab creates a differentiable, self-organizing virtual lab where agents can hypothesize, intervene, and learn mechanisms before touching wet labs, with benchmarks for recovery, reversal planning, open-ended discovery, and auditability.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 402