>>206237015>>206237598There's a model in stochastic dynamics called 'fluctuation theorem' which states that the average instantaneous entropy of the system must be positive, but depending on how wide the deviation of the entropy distribution is, there can be a non-zero probability of an instantaneous measurement having a negative entropy change on time scales that are small compared to the characteristic thermal timescale of the system.
There was a really neat research paper I read by a team at University of Iowa from a couple years ago where they used dust particles suspended in a plasma to demonstrate fluctuation theorem (the dynamics of the system are so slow that you can measure the instantaneous entropy). This is something similar the experiment isn't "traveling backwards through time", but under the right conditions they can see the instantaneous entropy of this system of entangled electrons briefly show a negative change.
tl;dr - it's not time travel, it's just a funny little quirk of statistical mechanics - the average entropy of a system has to increase over time, but systems can spontaneously experience negative changes on very short timescales without it violating the laws of thermodynamics
t - physicist
