Quantifying the relation between work and information - recent results using the smooth entropy approach

Date/Time: 03rd October 2011 / 15:00 - - ()

Speaker: Dr Oscar C.O. Dahlsten (University of Oxford)
Abstract

I will give the key points of two related papers: (1) arXiv:0908.0424, "The work value of information", and (2) arXiv:1009.1630, "The thermodynamic meaning of negative entropy". These papers are concerned with Landauer's principle, whereby information erasure costs work, as well as the inverse process wherein work is gained at the cost of losing information (Szilard's engine). (1) shows that one can quantify how much work one can get for a given amount of information in a quite general way using the recently developed smooth entropy approach (the von Neumann entropy turns out to be inadequate). (2) shows how to interpret negative conditional entropy as work GAIN from information erasure (again using the smooth entropy approach). Joint work with Aaberg, del Rio, Egloff, Renner, Rieper and Vedral.