SN-X
All research
PostJanuary 29, 2026

The Immune System as a Learning System

An essay review: we argue that adaptive immunity is best understood as a continual, embodied reinforcement-learning process operating on a population of B and T cells, with somatic hypermutation playing the role of a parameterised policy and affinity maturation as policy-gradient ascent.

SN-X Biology Group

Thesis

The adaptive immune system is a continual, embodied reinforcement learner. It observes (peptide–MHC presentation), acts (clonal expansion and differentiation), and updates (somatic hypermutation and selection).

Implications

  • Vaccine design becomes a problem in curriculum design.
  • Autoimmunity is a failure mode of the same learning algorithm that confers

protection — and so is intelligible in the same vocabulary.

  • The architecture of the immune system offers concrete predictions for the

design of continual-learning artificial agents.