Consider two extreme examples of uninteresting, unsurprising, boring data: A visionbased agent that always stays in the dark will experience an extremely compressible, soon totally predictable history of unchanging visual inputs. In front of a screen full of white noise conveying a lot of information and “novelty” and “surprise” in the traditional sense of Boltzmann and Shannon , however, it will experience highly unpredictable and fundamentally incompressible data. In both cases the data is boring as it does not allow for further compression progress. Therefore we reject the traditional notion of surprise. Neither the arbitrary nor the fully predictable is truly novel or surprising—only data with still unknown algorithmic regularities are
Juergen Schmidhuber - Driven by Compression Progress: A Simple Principle Explains Essential Aspects of Subjective Beauty, Novelty, Surprise, Interestingness, Attention, Curiosity, Creativity, Art, Science, Music, Jokes














