Lossy compression is confirmation bias, but that is only from a scientific perspective. Confirmation bias is not a dirty word/phrase from an engineering perspective. It is essential to utility -- unless one's utility function is the search for truth.
On Mon, Mar 30, 2020 at 6:48 AM TimTyler <t...@tt1.org> wrote: > On 2020-03-21 15:14:PM, Matt Mahoney wrote: > > > A lossless compression contest on video would result in contestants > > spending 99.9999% of their efforts on compressing data that the eye > > and brain throw away, assuming the payoff is the same for both types. > > Noise is not just white noise, but all the details in the scene that > > don't increase your reproductive fitness. > > Hmm. Disagree with Matt in his own domain. An intelligent > agent ought to be able to recognize at least some sources of > noise and not waste their time with them. Maybe a lot of the > losslessly-compressed data would represent noise, but that > is a bit different. It doesn't mean the agent spent a lot of > time fruitlessly trying to figure out how to compress it. > > If, for example the agent gets reward as it goes along for > how well it is going, it should spend its limited resources > where they count, which would involve recognizing - and > avoiding wasting time compressing - incompressible noise. > > -- > __________ > |im |yler http://timtyler.org/ > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T2a0cd9d392f9ff94-Mdd1fb519e0615f89188f77c0 Delivery options: https://agi.topicbox.com/groups/agi/subscription
