3 papers
cs.AI2026
Nociception as a Control Primitive: Afferent Channels and Nociceptive Memory for Agents Deployed in One Body
Wolfgang Maass
An agent deployed in a single body cannot learn how fast that body wears, because every trial that would reveal its wear resistance wears the body it would protect. We study this \…
cs.AI2026
Completion vs Optimality: Policy Gradient in Long-Horizon Cumulative-Damage Problems
Wolfgang Maass, Sabine Janzen
Long-horizon decision problems with cumulative damage couple locally attractive actions to globally adverse outcomes. We identify two orthogonal failure modes for policy-gradient m…
cs.LG2026
Evolving Afferent Architectures: Biologically-inspired Models for Damage-Avoidance Learning
Wolfgang Maass, Sabine Janzen, Prajvi Saxena +1
We introduce Afferent Learning, a framework that produces Computational Afferent Traces (CATs) as adaptive, internal risk signals for damage-avoidance learning. Inspired by biologi…