The name Kanerva rings a bell, the 1st edition of the RL book mentions "Kanerva coding" as an option for function approximation. What you tell sounds interestin

The name Kanerva rings a bell, the 1st edition of the RL book mentions "Kanerva coding" as an option for function approximation. What you tell sounds interesting, maybe I should look a bit deeper into his work.